Upload folder using huggingface_hub

.gitattributes

@@ -1,35 +1,41 @@
-*.7z filter=lfs diff=lfs merge=lfs -text
-*.arrow filter=lfs diff=lfs merge=lfs -text
-*.bin filter=lfs diff=lfs merge=lfs -text
-*.bz2 filter=lfs diff=lfs merge=lfs -text
-*.ckpt filter=lfs diff=lfs merge=lfs -text
-*.ftz filter=lfs diff=lfs merge=lfs -text
-*.gz filter=lfs diff=lfs merge=lfs -text
-*.h5 filter=lfs diff=lfs merge=lfs -text
-*.joblib filter=lfs diff=lfs merge=lfs -text
-*.lfs.* filter=lfs diff=lfs merge=lfs -text
-*.mlmodel filter=lfs diff=lfs merge=lfs -text
-*.model filter=lfs diff=lfs merge=lfs -text
-*.msgpack filter=lfs diff=lfs merge=lfs -text
-*.npy filter=lfs diff=lfs merge=lfs -text
-*.npz filter=lfs diff=lfs merge=lfs -text
-*.onnx filter=lfs diff=lfs merge=lfs -text
-*.ot filter=lfs diff=lfs merge=lfs -text
-*.parquet filter=lfs diff=lfs merge=lfs -text
-*.pb filter=lfs diff=lfs merge=lfs -text
-*.pickle filter=lfs diff=lfs merge=lfs -text
-*.pkl filter=lfs diff=lfs merge=lfs -text
-*.pt filter=lfs diff=lfs merge=lfs -text
-*.pth filter=lfs diff=lfs merge=lfs -text
-*.rar filter=lfs diff=lfs merge=lfs -text
-*.safetensors filter=lfs diff=lfs merge=lfs -text
-saved_model/**/* filter=lfs diff=lfs merge=lfs -text
-*.tar.* filter=lfs diff=lfs merge=lfs -text
-*.tar filter=lfs diff=lfs merge=lfs -text
-*.tflite filter=lfs diff=lfs merge=lfs -text
-*.tgz filter=lfs diff=lfs merge=lfs -text
-*.wasm filter=lfs diff=lfs merge=lfs -text
-*.xz filter=lfs diff=lfs merge=lfs -text
-*.zip filter=lfs diff=lfs merge=lfs -text
-*.zst filter=lfs diff=lfs merge=lfs -text
-*tfevents* filter=lfs diff=lfs merge=lfs -text
+*.large_file_extension filter=lfs diff=lfs merge=lfs -text
+*.mp4 filter=lfs diff=lfs merge=lfs -text
+data/databases/astro_physpace_json.db filter=lfs diff=lfs merge=lfs -text
+data/databases/astro_physpace_readable.db filter=lfs diff=lfs merge=lfs -text
+data/databases/bio_json.db filter=lfs diff=lfs merge=lfs -text
+data/databases/bio_readable.db filter=lfs diff=lfs merge=lfs -text
+data/databases/fluid_json.db filter=lfs diff=lfs merge=lfs -text
+data/databases/fluid_readable.db filter=lfs diff=lfs merge=lfs -text
+data/raw/astro_physpace/astro-ph_6000_last_updated_date_desc_20240715_134752.jsonl filter=lfs diff=lfs merge=lfs -text
+data/raw/evo_bio/q-bio.PE_10000_last_updated_date_desc_20240723_163228.jsonl filter=lfs diff=lfs merge=lfs -text
+data/raw/fluid_dyn/physics.flu-dyn_10000_last_updated_date_desc_20240723_155932.jsonl filter=lfs diff=lfs merge=lfs -text
+data/raw/phy_strat/stratified_arxiv_data_combined_astro_physpace.jsonl filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_json_QdeCg2hM_20240801_034716/logs/log.txt filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_json_QdeCg2hM_20240801_034716/predicted_responses.txt filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_json_QdeCg2hM_20240801_034716/predictions.json filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_json_QdeCg2hM_20240801_034716/prompts.txt filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_json_bsnlEItA_20240725_031746/logs/log.txt filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_json_bsnlEItA_20240725_031746/prompts.txt filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_json_combined_preds_across_runs_astrophyspace/new_random_json_NPhqPpXy_20240716_040707/logs/log.txt filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_json_combined_preds_across_runs_astrophyspace/new_random_json_lWUgWudS_20240717_164223/logs/log.txt filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_json_combined_preds_across_runs_astrophyspace/new_random_json_lWUgWudS_20240717_164223/prompts.txt filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_json_combined_preds_across_runs_astrophyspace/predictions.json filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_json_g1PPhwYp_20240725_031815/logs/log.txt filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_json_g1PPhwYp_20240725_031815/prompts.txt filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_readable_Y7zNEy1v_20240725_031826/duped_predictions.json filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_readable_Y7zNEy1v_20240725_031826/logs/log.txt filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_readable_Y7zNEy1v_20240725_031826/predicted_responses.txt filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_readable_Y7zNEy1v_20240725_031826/prompts.txt filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_readable_b1AKFDbd_20240725_031800/duped_predictions.json filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_readable_b1AKFDbd_20240725_031800/logs/log.txt filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_readable_b1AKFDbd_20240725_031800/predicted_responses.txt filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_readable_b1AKFDbd_20240725_031800/prompts.txt filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_readable_combined_preds_across_runs_astrophyspace/new_random_readable_RLKNTmuL_20240716_141932/prompts.txt filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_readable_combined_preds_across_runs_astrophyspace/new_random_readable_dpjyPtQ1_20240717_161329/duped_predictions.json filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_readable_combined_preds_across_runs_astrophyspace/new_random_readable_dpjyPtQ1_20240717_161329/logs/log.txt filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_readable_combined_preds_across_runs_astrophyspace/new_random_readable_dpjyPtQ1_20240717_161329/predicted_responses.txt filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_readable_combined_preds_across_runs_astrophyspace/new_random_readable_dpjyPtQ1_20240717_161329/prompts.txt filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_readable_combined_preds_across_runs_astrophyspace/predictions.json filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_readable_ffDyTFzv_20240801_034725/logs/log.txt filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_readable_ffDyTFzv_20240801_034725/predictions.json filter=lfs diff=lfs merge=lfs -text
+data/results/new_random_readable_ffDyTFzv_20240801_034725/prompts.txt filter=lfs diff=lfs merge=lfs -text

.github/workflows/update_space.yml

@@ -0,0 +1,28 @@
+name: Run Python script
+
+on:
+  push:
+    branches:
+      - main
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+
+    steps:
+    - name: Checkout
+      uses: actions/checkout@v2
+
+    - name: Set up Python
+      uses: actions/setup-python@v2
+      with:
+        python-version: '3.9'
+
+    - name: Install Gradio
+      run: python -m pip install gradio
+
+    - name: Log in to Hugging Face
+      run: python -c 'import huggingface_hub; huggingface_hub.login(token="${{ secrets.hf_token }}")'
+
+    - name: Deploy to Spaces
+      run: gradio deploy

.gitignore

@@ -0,0 +1,171 @@
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+cover/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+.pybuilder/
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+#   For a library or package, you might want to ignore these files since the code is
+#   intended to run in multiple environments; otherwise, check them in:
+# .python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# poetry
+#   Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#   https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
+#poetry.lock
+
+# pdm
+#   Similar to Pipfile.lock, it is generally recommended to include pdm.lock in version control.
+#pdm.lock
+#   pdm stores project-wide configurations in .pdm.toml, but it is recommended to not include it
+#   in version control.
+#   https://pdm.fming.dev/latest/usage/project/#working-with-version-control
+.pdm.toml
+.pdm-python
+.pdm-build/
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow and github.com/pdm-project/pdm
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# pytype static type analyzer
+.pytype/
+
+# Cython debug symbols
+cython_debug/
+
+# PyCharm
+#  JetBrains specific template is maintained in a separate JetBrains.gitignore that can
+#  be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
+#  and can be added to the global gitignore or merged into this file.  For a more nuclear
+#  option (not recommended) you can uncomment the following to ignore the entire idea folder.
+#.idea/
+main_batch.py
+requirements.txt
+wandb/
+slurm_logs/
+notebooks/
+assets/polymathic_data_files
+assets/notes
+assets/test.ipynb
+Meta-Llama-3-70B-Instruct/

.pre-commit-config.yaml

@@ -0,0 +1,6 @@
+repos:
+-   repo: https://github.com/psf/black
+    rev: 24.4.2
+    hooks:
+    - id: black
+      language_version: python3

LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2024 Abhipsha Das
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

README.md

@@ -1,12 +1,186 @@
 ---
-title: Surveyor 0
-emoji: 👀
-colorFrom: indigo
-colorTo: red
+title: surveyor-0
+app_file: scripts/run_db_interface.py
 sdk: gradio
-sdk_version: 4.44.0
-app_file: app.py
-pinned: false
+sdk_version: 4.40.0
 ---
+# Mapping the Data Landscape For Generalizable Scientific Models
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+We introduce a method to build a knowledge base to store structured information extracted from scientific publications, datasets and articles by leveraging large language models!
+
+We want to cover all of "science", and perform semantic search over scientific literature for highly specific knowledge discovery. 
+
+This tool helps us to find aggregate information and statistics pertaining to current state of scientific research, identify the gaps where current foundation models lack coverage and where they can generalize well, and helps discover overlaps of methods used across different fields, which can help facilitate in building more unified foundation models for science. 
+
+### Example Preview: Concept Co-occurrence Connectivity Graph for Astrophysics!
+
+
+
+https://github.com/user-attachments/assets/d0c2c4ac-924d-4ba8-80d5-5c665d910652
+
+
+
+We use the Llama-3-70B-Instruct model with 2 A100 80GB GPUs for structured information extraction. 
+
+## Workflow
+
+<table>
+  <tr>
+    <td width="50%" valign="top">
+      <img src="assets/eval_pipeline.png" alt="Fig 1" width="100%">
+      <p align="center">
+        <em>Fig 1: Prompt optimization pipeline to maximize precision of the model annotated 
+        predictions by running on manually annotated subset of scientific corpora. The 
+        tagged outputs can be generated as JSON or in a readable format, and be 
+        generated using temperature and nucleus sampling (sweep hyperparams).</em>
+      </p>
+    </td>
+    <td width="50%" valign="top">
+      <img src="assets/pipeline.png" alt="Fig 2" width="100%">
+      <p align="center">
+        <em>Fig 2: Illustration of the structured prediction pipeline on the full corpus of 
+        scientific papers, which runs optimized prompts and stores the model's outputs in 
+        a SQL db.</em>
+      </p>
+    </td>
+  </tr>
+</table>
+
+## Installation 
+
+Clone and navigate to the repository:
+```
+git clone https://github.com/chiral-carbon/kg-for-science.git
+cd kg-for-science
+```
+Create a conda or virtual environment:
+```
+conda env create -f environment.yml
+```
+Activate the environment:
+```
+conda activate kg4s
+```
+Set up code formatting and pre-commit hooks:
+```
+pre-commit install
+```
+## Quickstart
+
+### Run an existing DB
+
+To run an existing database in the `databases` directory, run SQLite in your terminal in the project root:
+
+```
+sqlite3 databases/<table_name>
+```
+
+### Launch a Gradio interface for SQL query search over the created databases
+```
+gradio scripts/run_db_interface.py
+```
+The interface shows all the created databases in the `data/databases` directory which can be loaded and queried.
+
+
+## Running the tool on new data 
+
+### Download raw data from arXiv
+
+Run `scripts/collect_data.py` to download papers for arXiv:
+```
+python scripts/collect_data.py --max_results 1000 --search_query astro-ph --sort_by 'last_updated_date' --sort_order desc --out_dir <dir name>
+```
+
+These are the default arguments, you can modify them to specify the arxiv channel, number of papers and order of search. 
+
+The data is stored in the `data/raw/<out_dir>` directory.
+The `out_dir` is a required argument that creates a new directory in `data/raw` and stores the scraped data in a jsonl file inside `data/raw/<out_dir>`. Refer to the [raw data README](data/raw/README.md) to see how the files are named. 
+
+### Schema and Annotations
+
+A schema was prepared by the authors of this project who were also the annotators for a small subset of the downloaded papers. 
+This schema defined tags to extract concepts from the downloaded scientific papers. They were used as reference by the annotators when manually creating a small subset of annotated papers. They were also passed as instructions to the language model to tag the papers. A set of consituency tests were also defined to resolve ambiguity and guide the annotation process. 
+
+The schema, tests and manual_annotations are stored in the `data/manual` directory. Refer to the [README on manual work](data/manual/README.md) done for the extraction process.
+
+
+### Run the model on downloaded arXiv raw data
+
+Run `main.py` to call Llama-3 70B Instruct and perform extractions on the downloaded papers from any of the `data/raw` folders using Slurm jobs:
+```
+sbatch run.sh
+```
+You can modify the arguments passed to `main.py` as required:
+- evaluation on dev set or extracting data with a new test dataset,
+- specifying the format of model responses (json or readable).
+
+You can view the options by running `python main.py --help`:
+```
+Usage: main.py [OPTIONS]
+
+Options:
+  --kind TEXT              Specify the kind of prompt input: json (default) or
+                           readable
+  --runtype [new|eval]     Specify the type of run: new or eval (default)
+  --data TEXT              Specify the directory of the data if running on new
+                           data
+  --sweep                  Run sweeps
+  --sweep_config TEXT      Sweep configuration file
+  --load_best_config TEXT  Load the best configuration from a file
+  --help                   Show this message and exit.
+```
+
+If bound by compute resources and unable to use the Llama-3-70B-Instruct model, you can choose a different model if limited by memory and GPU.  Since we use the Huggingface transformers library and the model hosted on Huggingface, any new model you want to load should be hosted there as well. 
+
+**Note:** In the eval mode when running on the dev set, the model was run for different sweeps for prompt optimization. The sweep details are stored in `sweep_config.json`.
+
+To load a different model, change the `DEFAULT_MODEL_PATH` value in `config.py` to the Huggingface `<repo_name>/<model_name>` that you want. 
+
+
+The current best performance on the dev set:
+
+| Metric | kind=json | kind=readable |
+|--------|-----------|---------------|
+| precision | 0.4329 | 0.4364 |
+| recall | 0.3974 | 0.3110 |
+| f1 | 0.4144 | 0.3632 |
+| union_precision | 0.5864 | 0.6242 |
+| union_recall | 0.5216 | 0.4459 |
+| union_f1 | 0.5521 | 0.5202 |
+| avg_time_per_sentence | 4.0315 | 2.7584 |
+| total_time | 463.6508 | 317.2468 |
+
+
+The processed data gets stored in `data/raw/results` under new directories named with arguments passed to `main.py`. Refer to the [results README](data/results/README.md) for inspecting the files that each directory stores and the naming convention.
+
+### Create a SQLite3 database of the predictions
+```
+python scripts/create_db.py --data_path <path to the jsonl file with data> --pred_path <path to the predictions.json file>
+```
+This creates a database in the `data/databases` directory. 
+```
+Usage: create_db.py [OPTIONS]
+
+Options:
+  --data_path TEXT         Path to the data file containing the papers
+                           information.
+  --pred_path TEXT         Path to the predictions file.
+  --db_name TEXT           Name of the database to create.
+  --force                  Force overwrite if database already exists
+  --help                   Show this message and exit.
+  ```
+
+All current databases are in the ```data/databases``` directory which can be downloaded and loaded with ```sqlite3``` to run queries on your own terminal. Refer to the [databases README](data/databases/README.md) for information on the tables that constitute each of the databases.
+
+
+## Relevant Resources for Reference
+### Tools
+- Nomic AI's Atlas has beautiful and interactive visualizations and also provides an embedding API for visualizing knowledge graphs | [Site](https://atlas.nomic.ai/) | [GitHub](https://github.com/nomic-ai/nomic)
+- Instagraph is a good starting point for KG viz. They generate the knowldge graph (nodes and edges) given a knowledge base using GPT 3.5 | [Site](https://instagraph.ai) | [GitHub](https://github.com/yoheinakajima/instagraph)
+- The Monarch Initiative has a neat interface for phenotype/gene/disease knowledge discovery | [Site](https://next.monarchinitiative.org) | [GitHub](https://github.com/monarch-initiative)
+- PaperQA: LLM Chain for answering questions from documents with citations | [GitHub](https://github.com/whitead/paper-qa)
+
+### Research Papers
+- Knowledge Graph in Astronomical Research with Large Language Models: Quantifying Driving Forces in Interdisciplinary Scientific Discovery | [arXiv](https://arxiv.org/pdf/2406.01391)
+- Graph of Thoughts: Solving Elaborate Problems with Large Language Models | [arXiv](https://arxiv.org/pdf/2308.09687)
+- Bring Your Own KG: Self-Supervised Program Synthesis for Zero-Shot KGQA | [arXiv](https://arxiv.org/pdf/2311.07850)

access_keys.json

@@ -0,0 +1 @@
+{"openai_api_key": "", "openai_org_id": "", "hf_token": "", "hf_token_write": ""}

assets/graph-prev.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:263c60b4a26b6d32e96ccdacd92a4edf9787581f5dee9b9eb182d8b1502044c0
+size 7540171

assets/graph.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28392771ef7d72e204adf2273cadc2728a6f89f9e54ab71c162bf387d596faa4
+size 21734557

assets/kg4s-demo.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f3faa731c6f5cb5bc694a56c80690acc0486356fd51bbe5163b4feff50d316f7
+size 90726535

assets/kg4s-graph.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef8d4980ac07ef08ea940fbc6ddb5c0fed460e385074be6eae1f069a9c7c4844
+size 27332577

assets/kg4s-preview.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:48078c4726cbea8dae008016bcafff51c40c16cd4d7914f8e01a16bac4102901
+size 8467193

config.py

@@ -0,0 +1,432 @@
+DEFAULT_MODEL_ID = "Meta-Llama-3-70B-Instruct"
+DEFAULT_INTERFACE_MODEL_ID = "NumbersStation/nsql-llama-2-7B"
+DEFAULT_KIND = "json"
+DEFAULT_TEMPERATURE = 0.6
+DEFAULT_TOP_P = 0.95
+DEFAULT_FEW_SHOT_NUM = 3
+DEFAULT_FEW_SHOT_SELECTION = "random"
+DEFAULT_SAVE_INTERVAL = 3
+DEFAULT_RES_DIR = "data/results"
+DEFAULT_LOG_DIR = "logs"
+DEFAULT_TABLES_DIR = "data/databases"
+
+canned_queries = [
+    (
+        "Modalities in Physics and Astronomy papers",
+        """
+     SELECT DISTINCT LOWER(concept) AS concept
+     FROM predictions
+     JOIN (
+         SELECT paper_id, url
+         FROM papers
+         WHERE primary_category LIKE '%physics.space-ph%'
+            OR primary_category LIKE '%astro-ph.%'
+     ) AS paper_ids
+     ON predictions.paper_id = paper_ids.paper_id 
+     WHERE predictions.tag_type = 'modality'
+     """,
+    ),
+    (
+        "Datasets in Evolutionary Biology that use PDEs",
+        """
+    WITH pde_predictions AS (
+    SELECT paper_id, concept AS pde_concept, tag_type AS pde_tag_type
+    FROM predictions
+    WHERE tag_type IN ('method', 'model')
+      AND (
+          LOWER(concept) LIKE '%pde%'
+          OR LOWER(concept) LIKE '%partial differential equation%'
+      )
+    )
+    SELECT DISTINCT 
+        papers.paper_id,
+        papers.url,
+        LOWER(p_dataset.concept) AS dataset,
+        pde_predictions.pde_concept AS pde_related_concept,
+        pde_predictions.pde_tag_type AS pde_related_type
+    FROM papers
+    JOIN pde_predictions ON papers.paper_id = pde_predictions.paper_id
+    LEFT JOIN predictions p_dataset ON papers.paper_id = p_dataset.paper_id
+    WHERE papers.primary_category LIKE '%q-bio.PE%'
+    AND (p_dataset.tag_type = 'dataset' OR p_dataset.tag_type IS NULL)
+    ORDER BY papers.paper_id, dataset, pde_related_concept;
+    """,
+    ),
+    (
+        "Trends in objects of study in Cosmology since 2019",
+        """
+    
+    SELECT 
+        substr(papers.updated_on, 2, 7) as year_month, 
+        predictions.concept as object, 
+        COUNT(DISTINCT papers.paper_id) as paper_count 
+    FROM 
+        papers 
+    JOIN 
+        predictions ON papers.paper_id = predictions.paper_id 
+    WHERE 
+        papers.primary_category LIKE '%astro-ph.CO%' 
+        AND predictions.tag_type = 'object' 
+        AND CAST(SUBSTR(papers.updated_on, 2, 4) AS INTEGER) >= 2019
+    GROUP BY 
+        year_month, object 
+    ORDER BY 
+        year_month DESC, paper_count DESC
+    """,
+    ),
+    (
+        "New datasets in fluid dynamics since 2020",
+        """
+    WITH ranked_datasets AS (
+    SELECT 
+        p.paper_id,         
+        p.url,
+        pred.concept AS dataset,
+        p.updated_on,
+        ROW_NUMBER() OVER (PARTITION BY pred.concept ORDER BY p.updated_on ASC) AS rn
+    FROM 
+        papers p
+    JOIN 
+        predictions pred ON p.paper_id = pred.paper_id
+    WHERE 
+        pred.tag_type = 'dataset'
+        AND p.primary_category LIKE '%physics.flu-dyn%'
+        AND CAST(SUBSTR(p.updated_on, 2, 4) AS INTEGER) >= 2020 
+    )
+    SELECT 
+        paper_id,
+        url,
+        dataset,
+        updated_on
+    FROM 
+        ranked_datasets
+    WHERE 
+        rn = 1
+    ORDER BY 
+        updated_on ASC    
+    """,
+    ),
+    (
+        "Evolutionary biology datasets that use spatiotemporal dynamics",
+        """
+    WITH evo_bio_papers AS (
+    SELECT paper_id
+    FROM papers
+    WHERE primary_category LIKE '%q-bio.PE%' 
+    ),
+    spatiotemporal_keywords AS (
+        SELECT 'spatio-temporal' AS keyword
+        UNION SELECT 'spatiotemporal'
+        UNION SELECT 'spatio-temporal'
+        UNION SELECT 'spatial and temporal'
+        UNION SELECT 'space-time'
+        UNION SELECT 'geographic distribution'
+        UNION SELECT 'phylogeograph'
+        UNION SELECT 'biogeograph'
+        UNION SELECT 'dispersal'
+        UNION SELECT 'migration'
+        UNION SELECT 'range expansion'
+        UNION SELECT 'population dynamics'
+    )
+    SELECT DISTINCT
+        p.paper_id,
+        p.updated_on,
+        p.abstract,
+        d.concept AS dataset,
+        GROUP_CONCAT(DISTINCT stk.keyword) AS spatiotemporal_keywords_found
+    FROM 
+        evo_bio_papers ebp
+    JOIN 
+        papers p ON ebp.paper_id = p.paper_id
+    JOIN 
+        predictions d ON p.paper_id = d.paper_id
+    JOIN 
+        predictions st ON p.paper_id = st.paper_id
+    JOIN 
+        spatiotemporal_keywords stk
+    WHERE 
+        d.tag_type = 'dataset'
+        AND st.tag_type = 'modality'
+        AND LOWER(st.concept) LIKE '%' || stk.keyword || '%'
+    GROUP BY 
+        p.paper_id, p.updated_on, p.abstract, d.concept
+    ORDER BY 
+        p.updated_on DESC
+    """,
+    ),
+    (
+        "What percentage of papers use only galaxy or spectra, or both or neither?",
+        """
+    WITH paper_modalities AS (
+    SELECT 
+        p.paper_id,
+        MAX(CASE WHEN LOWER(pred.concept) LIKE '%imag%' THEN 1 ELSE 0 END) AS uses_galaxy_images,
+        MAX(CASE WHEN LOWER(pred.concept) LIKE '%spectr%' THEN 1 ELSE 0 END) AS uses_spectra
+    FROM 
+        papers p
+    LEFT JOIN 
+        predictions pred ON p.paper_id = pred.paper_id
+    WHERE 
+            p.primary_category LIKE '%astro-ph%'
+            AND pred.tag_type = 'modality'
+        GROUP BY 
+            p.paper_id
+    ),
+    categorized_papers AS (
+        SELECT
+            CASE
+                WHEN uses_galaxy_images = 1 AND uses_spectra = 1 THEN 'Both'
+                WHEN uses_galaxy_images = 1 THEN 'Only Galaxy Images'
+                WHEN uses_spectra = 1 THEN 'Only Spectra'
+                ELSE 'Neither'
+            END AS category,
+            COUNT(*) AS paper_count
+        FROM 
+            paper_modalities
+        GROUP BY 
+            CASE
+                WHEN uses_galaxy_images = 1 AND uses_spectra = 1 THEN 'Both'
+                WHEN uses_galaxy_images = 1 THEN 'Only Galaxy Images'
+                WHEN uses_spectra = 1 THEN 'Only Spectra'
+                ELSE 'Neither'
+            END
+    )
+    SELECT
+        category,
+        paper_count,
+        ROUND(CAST(paper_count AS FLOAT) / (SELECT SUM(paper_count) FROM categorized_papers) * 100, 2) AS percentage
+    FROM
+        categorized_papers
+    ORDER BY
+        paper_count DESC
+    """,
+    ),
+    (
+        "What are all the next highest data modalities after images and spectra?",
+        """
+        SELECT 
+            LOWER(concept) AS modality,
+            COUNT(DISTINCT paper_id) AS usage_count
+        FROM 
+            predictions
+        WHERE 
+            tag_type = 'modality'
+            AND LOWER(concept) NOT LIKE '%imag%'
+            AND LOWER(concept) NOT LIKE '%spectr%'
+        GROUP BY 
+            LOWER(concept)
+        ORDER BY 
+            usage_count DESC
+        """,
+    ),
+    (
+        "If we include the next biggest data modality, how much does coverage change?",
+        """
+    WITH modality_counts AS (
+    SELECT 
+        LOWER(concept) AS modality,
+        COUNT(DISTINCT paper_id) AS usage_count
+    FROM 
+        predictions
+    WHERE 
+        tag_type = 'modality'
+        AND LOWER(concept) NOT LIKE '%imag%'
+        AND LOWER(concept) NOT LIKE '%spectr%'
+    GROUP BY 
+        LOWER(concept)
+    ORDER BY 
+        usage_count DESC
+    LIMIT 1
+    ),
+    paper_modalities AS (
+        SELECT 
+            p.paper_id,
+            MAX(CASE WHEN LOWER(pred.concept) LIKE '%imag%' THEN 1 ELSE 0 END) AS uses_galaxy_images,
+            MAX(CASE WHEN LOWER(pred.concept) LIKE '%spectr%' THEN 1 ELSE 0 END) AS uses_spectra,
+            MAX(CASE WHEN LOWER(pred.concept) LIKE (SELECT '%' || modality || '%' FROM modality_counts) THEN 1 ELSE 0 END) AS uses_third_modality
+        FROM 
+            papers p
+        LEFT JOIN 
+            predictions pred ON p.paper_id = pred.paper_id
+        WHERE 
+            p.primary_category LIKE '%astro-ph%'
+            AND pred.tag_type = 'modality'
+        GROUP BY 
+            p.paper_id
+    ),
+    coverage_before AS (
+        SELECT
+            SUM(CASE WHEN uses_galaxy_images = 1 OR uses_spectra = 1 THEN 1 ELSE 0 END) AS covered_papers,
+            COUNT(*) AS total_papers
+        FROM 
+            paper_modalities
+    ),
+    coverage_after AS (
+        SELECT
+            SUM(CASE WHEN uses_galaxy_images = 1 OR uses_spectra = 1 OR uses_third_modality = 1 THEN 1 ELSE 0 END) AS covered_papers,
+            COUNT(*) AS total_papers
+        FROM 
+            paper_modalities
+    )
+    SELECT
+        (SELECT modality FROM modality_counts) AS third_modality,
+        ROUND(CAST(covered_papers AS FLOAT) / total_papers * 100, 2) AS coverage_before_percent,
+        ROUND(CAST((SELECT covered_papers FROM coverage_after) AS FLOAT) / total_papers * 100, 2) AS coverage_after_percent,
+        ROUND(CAST((SELECT covered_papers FROM coverage_after) AS FLOAT) / total_papers * 100, 2) - 
+        ROUND(CAST(covered_papers AS FLOAT) / total_papers * 100, 2) AS coverage_increase_percent
+    FROM
+        coverage_before
+    """,
+    ),
+    (
+        "Coverage if we select the next 5 highest modalities?",
+        """
+    WITH ranked_modalities AS (
+    SELECT 
+        LOWER(concept) AS modality,
+        COUNT(DISTINCT paper_id) AS usage_count,
+        ROW_NUMBER() OVER (ORDER BY COUNT(DISTINCT paper_id) DESC) AS rank
+    FROM 
+        predictions
+    WHERE 
+        tag_type = 'modality'
+        AND LOWER(concept) NOT LIKE '%imag%'
+        AND LOWER(concept) NOT LIKE '%spectr%'
+    GROUP BY 
+        LOWER(concept)
+    ),
+    paper_modalities AS (
+        SELECT 
+            p.paper_id,
+            MAX(CASE WHEN LOWER(pred.concept) LIKE '%imag%' THEN 1 ELSE 0 END) AS uses_images,
+            MAX(CASE WHEN LOWER(pred.concept) LIKE '%spectr%' THEN 1 ELSE 0 END) AS uses_spectra,
+            MAX(CASE WHEN rm.rank = 1 THEN 1 ELSE 0 END) AS uses_modality_1,
+            MAX(CASE WHEN rm.rank = 2 THEN 1 ELSE 0 END) AS uses_modality_2,
+            MAX(CASE WHEN rm.rank = 3 THEN 1 ELSE 0 END) AS uses_modality_3,
+            MAX(CASE WHEN rm.rank = 4 THEN 1 ELSE 0 END) AS uses_modality_4,
+            MAX(CASE WHEN rm.rank = 5 THEN 1 ELSE 0 END) AS uses_modality_5
+        FROM 
+            papers p
+        LEFT JOIN 
+            predictions pred ON p.paper_id = pred.paper_id
+        LEFT JOIN
+            ranked_modalities rm ON LOWER(pred.concept) = rm.modality
+        WHERE 
+            p.primary_category LIKE '%astro-ph%'
+            AND pred.tag_type = 'modality'
+        GROUP BY 
+            p.paper_id
+    ),
+    cumulative_coverage AS (
+        SELECT
+            'Images and Spectra' AS modalities,
+            0 AS added_modality_rank,
+            SUM(CASE WHEN uses_images = 1 OR uses_spectra = 1 THEN 1 ELSE 0 END) AS covered_papers,
+            COUNT(*) AS total_papers
+        FROM 
+            paper_modalities
+        
+        UNION ALL
+        
+        SELECT
+            'Images, Spectra, and Modality 1' AS modalities,
+            1 AS added_modality_rank,
+            SUM(CASE WHEN uses_images = 1 OR uses_spectra = 1 OR uses_modality_1 = 1 THEN 1 ELSE 0 END) AS covered_papers,
+            COUNT(*) AS total_papers
+        FROM 
+            paper_modalities
+        
+        UNION ALL
+        
+        SELECT
+            'Images, Spectra, Modality 1, and 2' AS modalities,
+            2 AS added_modality_rank,
+            SUM(CASE WHEN uses_images = 1 OR uses_spectra = 1 OR uses_modality_1 = 1 OR uses_modality_2 = 1 THEN 1 ELSE 0 END) AS covered_papers,
+            COUNT(*) AS total_papers
+        FROM 
+            paper_modalities
+        
+        UNION ALL
+        
+        SELECT
+            'Images, Spectra, Modality 1, 2, and 3' AS modalities,
+            3 AS added_modality_rank,
+            SUM(CASE WHEN uses_images = 1 OR uses_spectra = 1 OR uses_modality_1 = 1 OR uses_modality_2 = 1 OR uses_modality_3 = 1 THEN 1 ELSE 0 END) AS covered_papers,
+            COUNT(*) AS total_papers
+        FROM 
+            paper_modalities
+        
+        UNION ALL
+        
+        SELECT
+            'Images, Spectra, Modality 1, 2, 3, and 4' AS modalities,
+            4 AS added_modality_rank,
+            SUM(CASE WHEN uses_images = 1 OR uses_spectra = 1 OR uses_modality_1 = 1 OR uses_modality_2 = 1 OR uses_modality_3 = 1 OR uses_modality_4 = 1 THEN 1 ELSE 0 END) AS covered_papers,
+            COUNT(*) AS total_papers
+        FROM 
+            paper_modalities
+        
+        UNION ALL
+        
+        SELECT
+            'Images, Spectra, Modality 1, 2, 3, 4, and 5' AS modalities,
+            5 AS added_modality_rank,
+            SUM(CASE WHEN uses_images = 1 OR uses_spectra = 1 OR uses_modality_1 = 1 OR uses_modality_2 = 1 OR uses_modality_3 = 1 OR uses_modality_4 = 1 OR uses_modality_5 = 1 THEN 1 ELSE 0 END) AS covered_papers,
+            COUNT(*) AS total_papers
+        FROM 
+            paper_modalities
+        )
+        SELECT
+            cc.modalities,
+            COALESCE(rm.modality, 'N/A') AS added_modality,
+            rm.usage_count AS added_modality_usage,
+            ROUND(CAST(cc.covered_papers AS FLOAT) / cc.total_papers * 100, 2) AS coverage_percent,
+            ROUND(CAST(cc.covered_papers AS FLOAT) / cc.total_papers * 100, 2) - 
+            LAG(ROUND(CAST(cc.covered_papers AS FLOAT) / cc.total_papers * 100, 2), 1, 0) OVER (ORDER BY cc.added_modality_rank) AS coverage_increase_percent
+        FROM
+            cumulative_coverage cc
+        LEFT JOIN
+            ranked_modalities rm ON cc.added_modality_rank = rm.rank
+        ORDER BY
+            cc.added_modality_rank
+    """,
+    ),
+    (
+        "List all papers",
+        "SELECT paper_id, abstract AS abstract_preview, authors, primary_category FROM papers",
+    ),
+    (
+        "Count papers by category",
+        "SELECT primary_category, COUNT(*) as paper_count FROM papers GROUP BY primary_category ORDER BY paper_count DESC",
+    ),
+    (
+        "Top authors with most papers",
+        """
+     WITH author_papers AS (
+         SELECT json_each.value AS author
+         FROM papers, json_each(papers.authors)
+     )
+     SELECT author, COUNT(*) as paper_count 
+     FROM author_papers 
+     GROUP BY author 
+     ORDER BY paper_count DESC
+     """,
+    ),
+    (
+        "Papers with 'quantum' in abstract",
+        "SELECT paper_id, abstract AS abstract_preview FROM papers WHERE abstract LIKE '%quantum%'",
+    ),
+    (
+        "Most common concepts",
+        "SELECT concept, COUNT(*) as concept_count FROM predictions GROUP BY concept ORDER BY concept_count DESC",
+    ),
+    (
+        "Papers with multiple authors",
+        """
+     SELECT paper_id, json_array_length(authors) as author_count, authors
+     FROM papers
+     WHERE json_array_length(authors) > 1
+     ORDER BY author_count DESC
+     """,
+    ),
+]

data/README.md

@@ -0,0 +1,35 @@
+This directory contains all the scraped, manually labeled, util and processed data for this project. 
+
+
+It contains of 4 folders: 
+```
+data
+├── manual
+├── raw
+├── results
+└── databases
+```
+
+------------------------------
+
+### `manual` 
+This folder contains the schema and constituency tests that were manually developed for efficiently labelling data.
+It also contains manual annotations of a small subset of the `raw` data using the defined schema and constituency tests.
+
+Check the [manual README](manual/README.md) for further details.
+
+### `raw`
+This folder contains the raw data that was scraped from arXiv using the [arxiv](https://pypi.org/project/arxiv/) Python wrapper for different channels of research.
+
+Check the [raw README](raw/README.md) for further details.
+
+### `results`
+This folder contains the processed results after running inference through the model. 
+The model generates predictions of tagged concepts from the raw data and stores under folders within this folder.
+
+Check the [results README](results/README.md) for further details.
+
+### `databases`
+This folder contains the SQL daatabases created for each pair of raw papers and the corresponding model predictions of tagged concepts. 
+
+Check the [databases README](databases/README.md) for further details.

data/databases/README.md

@@ -0,0 +1,32 @@
+- This folder contains all the SQL databases for the different processed data along with their raw data.
+
+- The databases are named after the arXiv category and the format of the generated data.
+
+Each file in this folder is a database containing 2 tables:
+- **papers**
+    
+    The papers data from the `raw` folder that was fed to the model.
+    
+    SCHEMA:
+    - paper_id TEXT PRIMARY KEY,
+    - abstract TEXT,
+    - authors TEXT,
+    - primary_category TEXT,
+    - url TEXT,
+    - updated_on TEXT,
+    - sentence_count INTEGER
+
+- **predictions**
+    
+    The corresponding model generations stored in the `results` folder.
+
+    SCHEMA:
+    - id INTEGER PRIMARY KEY AUTOINCREMENT,
+    - paper_id TEXT,
+    - sentence_index INTEGER,
+    - tag_type TEXT,
+    - concept TEXT,
+    - FOREIGN KEY (paper_id) REFERENCES papers(paper_id)
+
+
+To query any database, open SQLite in your terminal and specify the database name. 

data/databases/astro_physpace_json.db

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2babe5742fbfd808bfc6fca67e781f2b9cec8682c4205d24bd197989d2a1b991
+size 48816128

data/databases/astro_physpace_readable.db

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:58e91909bb5f96c17027e86c790764792f763e7d83e27fed0e0f1ca3b3a7e362
+size 44724224

data/databases/bio_json.db

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8c2bbd58121244b1fa380c8d7aecfa227658726bad0ab56af646cafeeb886441
+size 38563840

data/databases/bio_readable.db

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5bfe4c0e11b02a5eeef73f67e99b2ed901eb46b9f962576751ca8d8abd313121
+size 37539840

data/databases/fluid_json.db

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d36e32c1396b13e745f865ce3dcd2ef437a7f681da69d2995331908e4676296f
+size 40112128

data/databases/fluid_readable.db

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1de657c59ed50d9911518acd0888f481ca39c6823417e6ba24c371c77f6d8808
+size 43855872

data/manual/README.md

@@ -0,0 +1,37 @@
+```
+manual
+├── constituency_tests.json
+├── human_annotations.jsonl
+└── schema.json
+```
+
+----------------------------
+
+- `schema.json` contains the schema that informs the definition of the different concepts in a scientific article. 
+    
+    This schema was developed by the project's authors through a few iterations of discussion. 
+    
+    The schema was:
+    - Used to guide human annotators in labeling a subset of raw data
+    - passed to the language model with instructions for annotating the raw data during in-context learning
+
+
+    Defining the schema ensures consistency and standardization in identifying and categorizing scientific concepts for both humans and the language model. It provides a common framework for comparing and aggregating data across multiple papers.
+
+
+- `constituency_tests.json` contains the tests for constituency that were defined for clarifying any follow-up questions one may have about the schema and the information extraction/tagging process.
+
+    The constituency tests were also developed by the project's authors and is useful for:
+    - Ambiguity resolution during the manual annotations phase: We may encounter concepts that may be categorized as more than tag based on only the schema definitions
+    - Ambiguity resolution for the model while generating tagged concepts: Though not utilized for now, it can potentially be used as a chain-of-thought or reasoning tool and passed to the language model along with instructions in order to have refined, less ambiguous tag predictions
+
+    It is currently mainly used as a reference points for manual annotators when they encounter ambiguity. 
+
+
+- `human_annotations.jsonl` contains 20 manually annotated paper titles and abstracts from the `astro-ph` arXiv channel stored in json lines. 
+    
+    - They contain annotations done by the project's authors based on the defined schema. Constituency tests help in resolving ambiguity. 
+    - The annotations are used for prompt optimization by dividing it into `train` and `dev` sets:
+        - the first 3 papers are used as the `train` set for few-shot examples and the rest are used as the `dev` set and passed to the language model for generating tags
+        - across different sweeps of hyperparameters on the `train` set prefixes, the precision and recall of the predicted and grounth truth tags in the `dev` set are measured to identify the best hyperparameters
+    - These hyperparameters are then applied to the prefixes from the `train` set when generating tags for the `test` set--which is a large raw dataset from arXiv.

data/manual/constituency_tests.json

@@ -0,0 +1,45 @@
+{
+    "model": [
+        "Could you imagine it as the representation of a data system?",
+        "Could you use it to map observed data into an alternate representation?",
+        "Could you imagine it as a construct to simulate behavior of a system?",
+        "Examples: \n1. Autoregressive language modeling: Models language by predicting the next token \n2. Standard Model in physics"
+    ],
+    "task": [
+        "Is this a goal or an end unto itself?",
+        "Could you imagine multiple ways to accomplish this goal? If not, it might be a method.",
+        "Examples: \n1. Language modeling \n2. Physical surrogate modeling \n3. Predicting swarm dynamics"
+    ],
+    "dataset": [
+        "Could you imagine this as a set of factual information collected through observation, measurement, or research?",
+        "Examples: \n1. ATCA radio-continuum observations \n2. CIFAR-10 \n3. ATLAS experiment data"
+    ],
+    "field": [
+        "Could you imagine a conference, journal, or tutorial on this topic?",
+        "Examples: \n1. Cosmology \n2. Physics \n3. Machine Learning"
+    ],
+    "modality": [
+        "Could you imagine this as a way that some aspect/facet of factual information or data is represented?",
+        "Examples: \n1. Text is a modality where data is in the form of words and sentences. \n2. Images are a modality where data consists of pixel values arranged in a grid. \n3. Spectra are a modality that represent data as a series of intensity values across a range of wavelengths or frequencies."
+    ],
+    "method": [
+        "Do people use this as a way to accomplish a task?",
+        "Examples: \n1.Multiple Physics Pretraining \n2. Cross-modal contrastive learning \n3. Autoregressive training of Transformers \n4. Direct Preference Optimization \n5. Ordinary Differential Equations"
+    ],
+    "object": [
+        "Could you imagine it as a physical or abstract entity that can be examined and studied?",
+        "Examples: \n1. Galaxies \n2. Black holes \n3. Radio emission \n4. Transformers"
+    ],
+    "property": [
+        "Could you imagine it as an inherent attribute or characteristic of an entity that can be studied?",
+        "Could you imagine it as a quantitative measurement of an entity, method, data, or data modality?",
+        "Could you understand it as a concept that can describe an entity?",
+        "Does it make sense independently, without being attached to another tag? If so, it might not be a property.",
+        "Examples: \n1. Sampling frequency is a property of a signal \n2. Expansion rate is a property of supernovae \n3. Gaussianity is a property of a dataset or signal being normally distributed \n4. Scale invariance is a property of Einstein field equations \n5. Gravity or gravitational acceleration is a property of any object that has mass \n6. Mass is a property of all matter, like planets, stars or protons \n7. Magnetic field strength (B)"
+    ],
+    "instrument": [
+        "Can it be used to produce data?",
+        "Does it measure something?",
+        "Examples: \n1.HATNet \n2. Australia Telescope Compact Array (ATCA) \n3. Dark Energy Spectroscopic Instrument (DESI)"
+    ]
+}

data/manual/human_annotations.jsonl

@@ -0,0 +1,20 @@
+{"title": "Radio-Continuum Emission From The Young Galactic Supernova Remnant G1.9+0.3", "abstract": "We present an analysis of a new Australia Telescope Compact Array (ATCA) radio-continuum observation of supernova remnant (SNR) G1.9+0.3, which at an age of $\\sim$181$\\pm$25 years is the youngest known in the Galaxy. We analysed all available radio-continuum observations at 6-cm from the ATCA and the Very Large Array. Using this data we estimate an expansion rate for G1.9+0.3 of 0.563%$\\pm$0.078% per year between 1984 and 2009. We note that in the 1980's G1.9+0.3 expanded somewhat slower (0.484% per year) than more recently (0.641% per year). We estimate that the average spectral index between 20-cm and 6-cm, across the entire SNR is $\\alpha=-0.72\\pm 0.26$ which is typical for younger SNRs. At 6-cm, we detect an average of 6% fractionally polarised radio emission with a peak of 17%$\\pm$3%. The polarised emission follows the contours of the strongest of X-ray emission. Using the new equipartition formula we estimate a magnetic field strength of B$\\approx 273\\mu$G, which to date, is one of the highest magnetic field strength found for any SNR and consistent with G1.9+0.3 being a very young remnant. This magnetic field strength implies a minimum total energy of the synchrotron radiation of E$_{\\textrm{min}} \\approx$ 1.8$\\times$10$^{48}$ ergs.", "tagged_title": "<object>Radio-Continuum Emission</object> From The Young <object>Galactic Supernova Remnant G1.9+0.3</object>", "tagged_abstract": "We present an analysis of a new <dataset><instrument>Australia Telescope Compact Array (ATCA)</instrument> <modality>radio-continuum</modality> observation</dataset> of <object>supernova remnant (SNR) G1.9+0.3</object>, which at an <property>age</property> of $\\sim$181$\\pm$25 years is the youngest known in the <object>Galaxy</object>. We analysed all available <dataset><modality>radio-continuum</modality> observations</dataset at 6-cm from the <instrument>ATCA</instrument> and the <instrument>Very Large Array</instrument>. Using this data we estimate an <property>expansion rate</property> for <object>G1.9+0.3</object> of 0.563%$\\pm$0.078% per year between 1984 and 2009. We note that in the 1980's <object>G1.9+0.3</object> expanded somewhat slower (0.484% per year) than more recently (0.641% per year). We estimate that the <property>average spectral index</property> between 20-cm and 6-cm, across the entire <object>SNR</object> is $\\alpha=-0.72\\pm 0.26$ which is typical for younger <object>SNRs</object>. At 6-cm, we detect an average of 6% fractionally polarised <object>radio emission</object> with a peak of 17%$\\pm$3%. The <object>polarised emission</object> follows the contours of the strongest of <object>X-ray emission</object>. Using the new <method>equipartition formula</method> we estimate a <property>magnetic field strength</property> of B$\\approx 273\\mu$G, which to date, is one of the highest <property>magnetic field strength</property> found for any <object>SNR</object> and consistent with <object>G1.9+0.3</object> being a very young  <object>remnant</object>. This <property>magnetic field strength</property> implies a minimum total <property>energy</property> of the <object>synchrotron radiation</object> of E$_{\\textrm{min}} \\approx$ 1.8$\\times$10$^{48}$ ergs.", "tagged_phrases": {"object": ["Radio-Continuum Emission", "Galactic Supernova Remnant G1.9+0.3", "supernova remnant (SNR) G1.9+0.3", "Galaxy", "G1.9+0.3", "SNR", "SNRs", "radio emission", "polarised emission", "X-ray emission", "remnant", "synchrotron radiation"], "dataset": ["Australia Telescope Compact Array (ATCA) radio-continuum observation", "radio-continuum observations"], "instrument": ["Australia Telescope Compact Array (ATCA)", "Very Large Array"], "modality": ["radio-continuum"], "property": ["age", "expansion rate", "average spectral index", "magnetic field strength", "energy"], "method": ["equipartition formula"]}}
+{"title": "Trispectrum versus Bispectrum in Single-Field Inflation", "abstract": "In the standard slow-roll inflationary cosmology, quantum fluctuations in a single field, the inflaton, generate approximately Gaussian primordial density perturbations. At present, the bispectrum and trispectrum of the density perturbations have not been observed and the probability distribution for these perturbations is consistent with Gaussianity. However, Planck satellite data will bring a new level of precision to bear on this issue, and it is possible that evidence for non-Gaussian effects in the primordial distribution will be discovered. One possibility is that a trispectrum will be observed without evidence for a non-zero bispectrum. It is not difficult for this to occur in inflationary models where quantum fluctuations in a field other than the inflaton contribute to the density perturbations. A natural question to ask is whether such an observation would rule out the standard scenarios. We explore this issue and find that it is possible to construct single-field models in which inflaton-generated primordial density perturbations have an observable trispectrum, but a bispectrum that is too small to be observed by the Planck satellite. However, an awkward fine tuning seems to be unavoidable.", "tagged_title": "<property>Trispectrum</property> versus <property>Bispectrum</property> in <object>Single-Field Inflation</object>", "tagged_abstract": "In the standard <model>slow-roll inflationary <field>cosmology</field></model>, <object>quantum fluctuations</object> in a single <object>field</object>, the <object>inflaton</object>, generate approximately <property>Gaussian</property> <property>primordial</property> <object><modality>density</modality> perturbations</object>. At present, the <property>bispectrum</property> and <property>trispectrum</property> of the <object><modality>density</modality> perturbations</object> have not been observed and the probability distribution for these <object>perturbations</object> is consistent with <property>Gaussianity</property>. However, <dataset><instrument>Planck satellite</instrument> data</dataset> will bring a new level of precision to bear on this issue, and it is possible that evidence for <property>non-Gaussian</property> effects in the <object>primordial distribution</object> will be discovered. One possibility is that a <property>trispectrum</property> will be observed without evidence for a non-zero <property>bispectrum</property>. It is not difficult for this to occur in <model>inflationary models</model> where <object>quantum fluctuations</object> in a field other than the <object>inflaton</object> contribute to the <object><modality>density</modality> perturbations</object>. A natural question to ask is whether such an observation would rule out the standard scenarios. We explore this issue and find that it is possible to construct <model>single-field models</model> in which <property>inflaton-generated</property> <property>primordial</property> <object><modality>density</modality> perturbations</object> have an observable <property>trispectrum</property>, but a <property>bispectrum</property> that is too small to be observed by the <instrument>Planck satellite</instrument>. However, an awkward fine tuning seems to be unavoidable.", "tagged_phrases": {"property": ["Trispectrum", "Bispectrum", "Gaussian", "primordial", "bispectrum", "trispectrum", "Gaussianity", "non-Gaussian", "inflaton-generated"], "object": ["Single-Field Inflation", "quantum fluctuations", "field", "inflaton", "density perturbations", "perturbations", "primordial distribution"], "model": ["slow-roll inflationary cosmology", "inflationary models", "single-field models"], "field": ["cosmology"], "modality": ["density"], "dataset": ["Planck satellite data"], "instrument": ["Planck satellite"]}}
+{"title": "A differential study of the oxygen abundances in the Galactic thin and thick disks", "abstract": "First results from a study into the abundance trends of oxygen in the Galactic thin and thick disks are presented. Oxygen abundances for 63 nearby F and G dwarf stars, based on very high resolution spectra ($R\\sim215 000$) and high signal-to-noise ($S/N>400$) of the faint forbidden oxygen line at 6300 {\\AA}, have been determined. Our findings can be summarized as follows: {\\bf 1)} at $\\rm [Fe/H]<0$ the oxygen trends in the thin and thick disk are smooth and distinct, indicating their different origins, {\\bf 2)} $\\rm [O/Fe]$ for the thick disk stars show a turn-over at [Fe/H]$\\sim -0.35$, indicating the peak of the enrichment from SNe type Ia to the interstellar medium, {\\bf 3)} the thin disk stars show a shallow decrease going from $\\rm [Fe/H] \\sim -0.7$ to the highest metallicities with no apparent turn-over present indicating a more quiet star formation history, and {\\bf 4)} $\\rm [O/Fe]$ continues to decrease at $\\rm [Fe/H]>0$ without showing the leveling out that previously has been seen.", "tagged_title": "A differential study of the <object><object>oxygen</object> abundances</object> in the Galactic <property>thin</property> and <property>thick</property> <object>disks</object>", "tagged_abstract": "First results from a study into the <property>abundance trends</property> of <object>oxygen</object> in the Galactic <property>thin</property> and <property>thick</property> <object>disks</object> are presented. <object><object>Oxygen</object> abundances</object> for 63 nearby <object>F</object> and <object>G</object> <object>dwarf stars</object>, based on very high resolution <modality>spectra</modality> (R~215000) and high <property>signal-to-noise</property> (S/N>400) of the faint forbidden <object>oxygen line</object> at 6300 \u00c5, have been determined. Our findings can be summarized as follows: {\\bf 1)} at [Fe/H] < 0 the <object><object>oxygen</object> trends</object> in the <property>thin</property> and <property>thick</property> <object>disk</object> are smooth and distinct, indicating their different origins, {\\bf 2)} [O/Fe] for the <object><property>thick disk</property> stars</object> show a turn-over at [Fe/H] ~ \u22120.35, indicating the peak of the enrichment from SNe type Ia to the <object>interstellar medium</object>, {\\bf 3)} the <object><property>thin disk</property> stars</object> show a shallow decrease going from [Fe/H] ~ \u22120.7 to the highest <property>metallicities</property> with no apparent turn-over present indicating a more quiet <object>star</object> formation history, and {\\bf 4)} [O/Fe] continues to decrease at [Fe/H] > 0 without showing the leveling out that previously has been seen.", "tagged_phrases": {"object": ["oxygen abundances", "oxygen", "disks", "Oxygen abundances", "Oxygen", "F", "G", "dwarf stars", "oxygen line", "oxygen trends", "disk", "thick disk stars", "interstellar medium", "thin disk stars", "star"], "property": ["thin", "thick", "abundance trends", "signal-to-noise", "thick disk", "thin disk", "metallicities"], "modality": ["spectra"]}}
+{"title": "The radiative transfer at second order: a full treatment of the Boltzmann equation with polarization", "abstract": "This article investigates the full Boltzmann equation up to second order in the cosmological perturbations. Describing the distribution of polarized radiation by a tensor valued distribution function, we study the gauge dependence of the distribution function and summarize the construction of the gauge-invariant distribution function. The Liouville operator which describes the free streaming of electrons, and the collision term which describes the scattering of photons on free electrons are computed up to second order. Finally, the remaining dependence in the direction of the photon momentum is handled by expanding in projected symmetric trace-free multipoles and also in the more commonly used normal modes components. The results obtained remain to be used for computing numerically the contribution in the cosmic microwave background bispectrum which arises from the evolution of second order perturbations, in order to disentangle the primordial non-Gaussianity from the one generated by the subsequent non-linear evolution.", "tagged_title": "The <object>radiative transfer at second order</object>: a full treatment of the <method>Boltzmann equation</method> with polarization", "tagged_abstract": "This article investigates the full <method>Boltzmann equation</method> up to second order in the <object><field>cosmological</field> perturbations</object>. Describing the distribution of <object>polarized radiation</object> by a <method>tensor valued distribution function</method>, we study the <property>gauge dependence</property> of the <method>distribution function</method> and summarize the construction of the <method><property>gauge-invariant</property> distribution function</method>. The <model>Liouville operator</model> which describes the free streaming of  <object>electrons</object>, and the <object>collision term</object> which describes the <object>scattering of photons</object> on <object>free electrons</object> are computed up to second order. Finally, the remaining <property>dependence</property> in the direction of the <object>photon</object> <property>momentum</property> is handled by expanding in projected <property>symmetric</property> <property>trace-free</property> <object>multipoles</object> and also in the more commonly used <object>normal modes components</object>. The results obtained remain to be used for computing numerically the contribution in the <object>cosmic microwave background <property>bispectrum</property></object> which arises from the evolution of <object>second order perturbations</object>, in order to disentangle the <property>primordial non-Gaussianity</property> from the one generated by the subsequent <object><property>non-linear</property> evolution</object>.", "tagged_phrases": {"object": ["radiative transfer at second order", "cosmological perturbations", "polarized radiation", "electrons", "collision term", "scattering of photons", "free electrons", "photon", "multipoles", "normal modes components", "cosmic microwave background bispectrum", "second order perturbations", "non-linear evolution"], "method": ["Boltzmann equation", "tensor valued distribution function", "distribution function", "gauge-invariant distribution function"], "field": ["cosmological"], "property": ["gauge dependence", "gauge-invariant", "dependence", "momentum", "symmetric", "trace-free", "bispectrum", "primordial non-Gaussianity", "non-linear"], "model": ["Liouville operator"]}}
+{"title": "Independent confirmation and refined parameters of the hot Jupiter XO-5b", "abstract": "We present HATNet observations of XO-5b, confirming its planetary nature based on evidence beyond that described in the announcement of Burke et al. (2008), namely, the lack of significant correlation between spectral bisector variations and orbital phase. In addition, using extensive spectroscopic measurements spanning multiple seasons, we investigate the relatively large scatter in the spectral line bisectors. We also examine possible blended stellar configurations (hierarchical triples, chance alignments) that can mimic the planet signals, and we are able to show that none are consistent with the sum of all the data. The analysis of the S activity index shows no significant stellar activity. Our results for the planet parameters are consistent with values in Burke et al. (2008), and we refine both the stellar and planetary parameters using our data. XO-5b orbits a slightly evolved, late G type star with mass M_s = 0.88 +/- 0.03, radius R_s = 1.08 +/- 0.04, and metallicity close to solar. The planetary mass and radius are M_p = 1.059 +/- 0.028 M_Jup and R_p = 1.109 +/- 0.050 R_Jup, respectively, corresponding to a mean density of 0.96 -0.11 +0.14 g/cm^3. The ephemeris for the orbit is P = 4.187757 +/- 0.000011, E= 2454552.67168 +/- 0.00029 (BJD) with transit duration of 0.1307 +/- 0.0013 d. By measuring four individual transit centers, we found no signs for transit timing variations. The planet XO-5b is notable for its anomalously high Safronov number, and has a high surface gravity when compared to other transiting exoplanets with similar period.", "tagged_title": "Independent confirmation and refined parameters of the hot <object>Jupiter XO-5b</object>", "tagged_abstract": "We present <dataset><instrument>HATNet</instrument> observations</dataset> of <object>XO-5b</object>, confirming its planetary nature based on evidence beyond that described in the announcement of Burke et al. (2008), namely, the lack of significant <property>correlation</property> between <property>spectral bisector variations</property> and <property>orbital phase</property>. In addition, using extensive <modality>spectroscopic measurements</modality> spanning multiple seasons, we investigate the relatively large <property>scatter</property> in the <property>spectral line bisectors</property>. We also examine possible blended <object>stellar configurations</object> (<object>hierarchical triples</object>, <object>chance alignments</object>) that can mimic the <object>planet signals</object>, and we are able to show that none are consistent with the sum of all the data. The analysis of the <property>S activity index</property> shows no significant <object>stellar activity</object>. Our results for the <object>planet parameters</object> are consistent with values in Burke et al. (2008), and we refine both the <object>stellar and planetary parameters</object> using our data. <object>XO-5b</object> orbits a slightly evolved, late <object>G type star</object> with <property>mass</property> <property>M_s</property> = 0.88 +/- 0.03, <property>radius</property> <property>R_s</property> = 1.08 +/- 0.04, and <property>metallicity</property> close to solar. The planetary <property>mass</property> and <property>radius</property> are <property>M_p</property> = 1.059 +/- 0.028 M_Jup and <property>R_p</property> = 1.109 +/- 0.050 R_Jup, respectively, corresponding to a <property>mean density</property> of 0.96 -0.11 +0.14 g/cm^3. The <property>ephemeris</property> for the <property>orbit</property> is <property>P</property> = 4.187757 +/- 0.000011, <property>E</property> = 2454552.67168 +/- 0.00029 (BJD) with <property>transit duration</property> of 0.1307 +/- 0.0013 d. By measuring four individual transit centers, we found no signs for <object>transit timing variations</object>. The planet <object>XO-5b</object> is notable for its anomalously high <property>Safronov number</property>, and has a high surface <property>gravity</property> when compared to other transiting <object>exoplanets</object> with similar period.", "tagged_phrases": {"object": ["Jupiter XO-5b", "XO-5b", "stellar configurations", "hierarchical triples", "chance alignments", "planet signals", "stellar activity", "planet parameters", "stellar and planetary parameters", "G type star", "transit timing variations", "exoplanets"], "dataset": ["HATNet observations"], "instrument": ["HATNet"], "property": ["correlation", "spectral bisector variations", "orbital phase", "scatter", "spectral line bisectors", "S activity index", "mass", "M_s", "radius", "R_s", "metallicity", "M_p", "R_p", "mean density", "ephemeris", "orbit", "P", "E", "transit duration", "Safronov number", "gravity"], "modality": ["spectroscopic measurements"]}}
+{"title": "Electron Temperatures in W51 Complex from High Resolution, Low Frequency Radio Observations", "abstract": "W51 is a giant radio complex lying along the tangent to the Sagitarius arm at a distance of about 7kpc from Sun, with an extension of about 1deg in the sky. It is divided into three components A,B,C where W51A and W51B consist of many compact HII regions while W51C is a supernova remnant. We have made continuum radio observations of these HII regions of the W51 complex at 240,610,1060,1400 MHz using GMRT with lower resolution 20''x15'' at the lowest frequency. The observed spectra of the prominent thermal subcomponents of W51 have been fitted to a free-free emission spectrum and their physical properties like electron temperatures and emission measures have been estimated. The electron temperatures from continuum spectra are found to be lower than the temperatures reported from radio recombination line (RRL) studies of these HII regions indicating the need for a filling factor even at this resolution. Also, the observed brightness at 240MHz is found to be higher than expected from the best fits suggesting the need for a multicomponent model for the region.", "tagged_title": "<property>Electron Temperatures</property> in <object>W51 Complex</object> from <property>High Resolution</property>, <property>Low Frequency</property> <modality>Radio</modality> Observations", "tagged_abstract": "<object>W51</object> is a giant <object><modality>radio</modality> complex</object> lying along the tangent to the  <object>Sagitarius arm</object> at a distance of about 7kpc from <object>Sun</object>, with an extension of about 1deg in the sky. It is divided into three components A,B,C where <object>W51A</object> and <object>W51B</object> consist of many compact <object>HII regions </object> while <object>W51C</object> is a <object>supernova remnant</object>. We have made <dataset><modality>continuum radio</modality> observations</dataset> of these <object>HII regions</object> of the <object>W51</object> complex at 240,610,1060,1400 MHz using <instrument>GMRT</instrument> with lower resolution 20''x15'' at the lowest frequency. The observed <modality>spectra</modality> of the prominent <property>thermal subcomponents</property> of <object>W51</object> have been fitted to a free-free <object>emission spectrum</object> and their physical properties like <property>electron temperatures</property> and <property>emission measures</property> have been estimated. The <property>electron temperatures</property> from <modality>continuum spectra</modality> are found to be lower than the <property>temperatures</property> reported from <method>radio recombination line (RRL) studies</method> of these <object>HII regions</object> indicating the need for a filling factor even at this resolution. Also, the observed <property>brightness</property> at 240MHz is found to be higher than expected from the best fits suggesting the need for a <model>multicomponent model</model> for the region.", "tagged_phrases": {"property": ["Electron Temperatures", "High Resolution", "Low Frequency", "thermal subcomponents", "electron temperatures", "emission measures", "temperatures", "brightness"], "object": ["W51 Complex", "W51", "radio complex", "Sagitarius arm", "Sun", "W51A", "W51B", "HII regions", "W51C", "supernova remnant", "emission spectrum"], "modality": ["Radio", "radio", "continuum radio", "spectra", "continuum spectra"], "dataset": ["continuum radio observations"], "instrument": ["GMRT"], "method": ["radio recombination line (RRL) studies"], "model": ["multicomponent model"]}}
+{"title": "Dynamical Study of A1689 from Wide-Field VLT/VIMOS Spectroscopy: Mass Profile, Concentration Parameter, and Velocity Anisotropy", "abstract": "We examine the dynamics structure of the rich cluster A1689, combining VLT/VIMOS spectroscopy with Subaru/Suprime-Cam imaging. The radial velocity distribution of $\\sim 500$ cluster members is bounded by a pair of clearly defined velocity caustics, with a maximum amplitude of $\\sim|4000|$ km/s at $\\simeq$ 300 h$^{-1}$ kpc, beyond which the amplitude steadily declines, approaching zero velocity at a limiting radius of $\\sim$ 2 h$^{-1}$ Mpc. We derive the 3D velocity anisotropy and galaxy number density profiles using a model-independent method to solve the Jeans equation, simultaneously incorporating the observed velocity dispersion profile, the galaxy counts from deep Subaru imaging, and our previously derived cluster mass profile from a joint lensing and X-ray analysis. The velocity anisotropy is found to be predominantly radial at large radius, becoming increasingly tangential towards the center, in accord with expectations. We also analyze the galaxy data independently of our previous analysis using two different methods: The first is based on a solution of the Jeans equation assuming an NFW form for the mass distribution, whereas in the second method the caustic amplitude is used to determine the escape velocity. The cluster virial mass derived by both of these dynamical methods is in good agreement with results from our earlier lensing and X-ray analysis. We also confirm the high NFW concentration parameter, with results from both methods combined to yield $c_{\\rm vir}>13$ (1$\\sigma$). The inferred virial radius is consistent with the limiting radius where the caustics approach zero velocity and where the counts of cluster members drop off, suggesting that infall onto A1689 is currently not significant.", "tagged_title": "Dynamical Study of <object>A1689</object> from <object>Wide-Field <instrument>VLT/VIMOS</instrument> <modality>Spectroscopy</modality></object>: <property>Mass Profile</property>, <property>Concentration Parameter</property>, and <property>Velocity Anisotropy</property>", "tagged_abstract": "We examine the dynamics structure of the rich <object>cluster A1689</object>, combining <modality><instrument>VLT/VIMOS</instrument> spectroscopy</modality> with <modality><instrument>Subaru/Suprime-Cam</instrument> imaging</modality>. The <property>radial velocity distribution</property> of $\\sim 500$ <object>cluster members</object> is bounded by a pair of clearly defined velocity caustics, with a maximum <property>amplitude</property> of $\\sim|4000|$ km/s at $\\simeq$ 300 h$^{-1}$ kpc, beyond which the <property>amplitude</property> steadily declines, approaching zero velocity at a limiting <property>radius</property> of $\\sim$ 2 h$^{-1}$ Mpc. We derive the <property>3D velocity anisotropy</property> and <property>galaxy number density</property> profiles using a model-independent method to solve the <method>Jeans equation</method>, simultaneously incorporating the observed <property>velocity dispersion</property> profile, the <property>galaxy counts</property> from deep <instrument>Subaru imaging</instrument>, and our previously derived <property>cluster mass</property> profile from a <modality>joint lensing</modality> and <modality>X-ray</modality> analysis. The <property>velocity anisotropy</property> is found to be predominantly <property>radial</property> at large <property>radius</property>, becoming increasingly tangential towards the center, in accord with expectations. We also analyze the <object>galaxy</object> data independently of our previous analysis using two different methods: The first is based on a solution of the <method>Jeans equation</method> assuming an <method>NFW form for the mass distribution</method>, whereas in the second method the <property>caustic amplitude</property> is used to determine the <property>escape velocity</property>. The <object>cluster</object> <property>virial mass</property> derived by both of these dynamical methods is in good agreement with results from our earlier <modality>lensing</modality> and <modality>X-ray</modality> analysis. We also confirm the high <property>NFW concentration parameter</property>, with results from both methods combined to yield $c_{\\rm vir}>13$ (1$\\sigma$). The inferred <property>virial radius</property> is consistent with the limiting <property>radius</property> where the <property>caustics</property> approach zero <property>velocity</property> and where the counts of <object>cluster</object> members drop off, suggesting that infall onto <object>A1689</object> is currently not significant.", "tagged_phrases": {"object": ["A1689", "Wide-Field VLT/VIMOS Spectroscopy", "cluster A1689", "cluster members", "galaxy", "cluster"], "instrument": ["VLT/VIMOS", "Subaru/Suprime-Cam", "Subaru imaging"], "modality": ["Spectroscopy", "VLT/VIMOS spectroscopy", "Subaru/Suprime-Cam imaging", "joint lensing", "X-ray", "lensing"], "property": ["Mass Profile", "Concentration Parameter", "Velocity Anisotropy", "radial velocity distribution", "amplitude", "radius", "3D velocity anisotropy", "galaxy number density", "velocity dispersion", "galaxy counts", "cluster mass", "velocity anisotropy", "radial", "caustic amplitude", "escape velocity", "virial mass", "NFW concentration parameter", "virial radius", "caustics", "velocity"], "method": ["Jeans equation", "NFW form for the mass distribution"]}}
+{"title": "Systematic thermal reduction of neutronization in core-collapse supernovae", "abstract": "We investigate to what extent the temperature dependence of the nuclear symmetry energy can affect the neutronization of the stellar core prior to neutrino trapping during gravitational collapse. To this end, we implement a one-zone simulation to follow the collapse until beta equilibrium is reached and the lepton fraction remains constant. Since the strength of electron capture on the neutron-rich nuclei associated to the supernova scenario is still an open issue, we keep it as a free parameter. We find that the temperature dependence of the symmetry energy consistently yields a small reduction of deleptonization, which corresponds to a systematic effect on the shock wave energetics: the gain in dissociation energy of the shock has a small yet non-negligible value of about 0.4 foe (1 foe = 10^51 erg) and this result is almost independent from the strength of nuclear electron capture. The presence of such a systematic effect and its robustness under changes of the parameters of the one-zone model are significative enough to justify further investigations with detailed numerical simulations of supernova explosions.", "tagged_title": "Systematic thermal reduction of <property>neutronization</property> in <property>core-collapse</property> <object>supernovae</object>", "tagged_abstract": "We investigate to what extent the <property>temperature dependence</property> of the <object>nuclear symmetry energy</object> can affect the <property>neutronization</property> of the <object>stellar core</object> prior to <object>neutrino</object> trapping during <object>gravitational collapse</object>. To this end, we implement a <method>one-zone simulation</method> to follow the <property>collapse</property> until <property>beta equilibrium</property> is reached and the <object>lepton fraction</object> remains constant. Since the <property>strength</property> of <object><object>electron</object> capture</object> on the <property>neutron-rich</property> <object>nuclei</object> associated to the <object>supernova</object> scenario is still an open issue, we keep it as a free parameter. We find that the <property>temperature dependence</property> of the <object>symmetry energy</object> consistently yields a small reduction of <object>deleptonization</object>, which corresponds to a systematic effect on the <object>shock wave energetics</object>: the gain in <property>dissociation energy</property> of the shock has a small yet non-negligible value of about 0.4 foe (1 foe = 10^51 erg) and this result is almost independent from the <property> strength</property> of <object>nuclear <object>electron</object> capture</object>. The presence of such a systematic effect and its robustness under changes of the parameters of the <method>one-zone model</method> are significative enough to justify further investigations with detailed <task>numerical simulations</task> of <object>supernova explosions</object>.", "tagged_phrases": {"property": ["neutronization", "core-collapse", "temperature dependence", "collapse", "beta equilibrium", "strength", "neutron-rich", "dissociation energy"], "object": ["supernovae", "nuclear symmetry energy", "stellar core", "neutrino", "gravitational collapse", "lepton fraction", "electron capture", "electron", "nuclei", "supernova", "symmetry energy", "deleptonization", "shock wave energetics", "nuclear electron capture", "supernova explosions"], "method": ["one-zone simulation", "one-zone model"], "task": ["numerical simulations"]}}
+{"title": "Hadrons As Kerr-Newman Black Holes", "abstract": "The scale invariance of the source-free Einstein field equations suggests that one might be able to model hadrons as \"strong gravity\" black holes, if one uses an appropriate rescaling of units or a revised gravitational coupling factor. The inner consistency of this hypothesis is tested by retrodicting a close approximation to the mass of the proton from an equation that relates the angular momentum and mass of a Kerr black hole. More accurate mass and radius values for the proton are then retrodicted using the geometrodynamics form of the full Kerr-Newman solution of the Einstein-Maxwell equations. The radius of an alpha particle is calculated as an additional retrodictive test. In a third retrodictive test of the \"strong gravity\" hypothesis, the subatomic particle mass spectrum in the 100 MeV to 7,000 MeV range is retrodicted to a first approximation using the Kerr solution of General Relativity. The particle masses appear to form a restricted set of quantized values of the Kerr solution: n^1/2 M, where values of n are a set of discrete integers and M is the revised Planck mass. The accuracy of the 27 retrodicted masses averages 98.4%. Finally, the new atomic scale gravitational coupling constant suggests a radical revision of the assumptions governing the Planck scale, and leads to a natural explanation for the fine structure constant.", "tagged_title": "<object>Hadrons</object> As <object>Kerr-Newman Black Holes</object>", "tagged_abstract": "The <property>scale invariance</property> of the source-free <model>Einstein field equations</model> suggests that one might be able to model <object>hadrons</object> as <property>\"strong gravity\"</property> <object>black holes</object>, if one uses an appropriate rescaling of units or a revised <object>gravitational coupling factor</object>. The inner consistency of this hypothesis is tested by <task>retrodicting</task> a close approximation to the <property>mass</property> of the <object>proton</object> from an equation that relates the <property>angular momentum</property> and <property>mass</property> of a <object>Kerr black hole</object>. More accurate <property>mass</property> and <property>radius</property> values for the <object>proton</object> are then retrodicted using the <method>geometrodynamics form</method> of the <method>full Kerr-Newman solution</method> of the <model>Einstein-Maxwell equations</model>. The <property>radius</property> of an <object>alpha particle</object> is calculated as an additional <task>retrodictive test</task>. In a third <task>retrodictive test</task> of the \"strong gravity\" hypothesis, the <modality>subatomic particle mass spectrum</modality> in the 100 MeV to 7,000 MeV range is <task>retrodicted</task> to a first approximation using the <method>Kerr solution of General Relativity</method>. The <object>particle masses</object> appear to form a restricted set of quantized values of the <method>Kerr solution</method>: n^1/2 M, where values of <property>n</property> are a set of discrete integers and <property>M</property> is the revised <property>Planck mass</property>. The accuracy of the 27 <object>retrodicted masses</object> averages 98.4%. Finally, the new atomic scale <object>gravitational coupling constant</object> suggests a radical revision of the assumptions governing the <object>Planck scale</object>, and leads to a natural explanation for the <object>fine structure constant</object>.", "tagged_phrases": {"object": ["Hadrons", "Kerr-Newman Black Holes", "hadrons", "black holes", "gravitational coupling factor", "proton", "Kerr black hole", "alpha particle", "particle masses", "retrodicted masses", "gravitational coupling constant", "Planck scale", "fine structure constant"], "property": ["scale invariance", "\"strong gravity\"", "mass", "angular momentum", "radius", "n", "M", "Planck mass"], "model": ["Einstein field equations", "Einstein-Maxwell equations"], "task": ["retrodicting", "retrodictive test", "retrodicted"], "method": ["geometrodynamics form", "full Kerr-Newman solution", "Kerr solution of General Relativity", "Kerr solution"], "modality": ["subatomic particle mass spectrum"]}}
+{"title": "Large-Scale Magnetic-Field Generation by Randomly Forced Shearing Waves", "abstract": "A rigorous theory for the generation of a large-scale magnetic field by random non-helically forced motions of a conducting fluid combined with a linear shear is presented in the analytically tractable limit of low Rm and weak shear. The dynamo is kinematic and due to fluctuations in the net (volume-averaged) electromotive force. This is a minimal proof-of-concept quasilinear calculation aiming to put the shear dynamo, a new effect recently found in numerical experiments, on a firm theoretical footing. Numerically observed scalings of the wavenumber and growth rate of the fastest growing mode, previously not understood, are derived analytically. The simplicity of the model suggests that shear dynamo action may be a generic property of sheared magnetohydrodynamic turbulence.", "tagged_title": "<task>Large-Scale Magnetic-Field Generation</task> by <method>Randomly Forced Shearing Waves</method>", "tagged_abstract": "A rigorous theory for the <task>generation of a large-scale <object>magnetic field</object></task> by <method>random non-helically forced motions</method> of a <object>conducting fluid</object> combined with a <method>linear shear</method> is presented in the analytically tractable limit of low Rm and weak shear. The <property>dynamo</property> is <property>kinematic</property> and due to fluctuations in the net (volume-averaged) <object>electromotive force</object>. This is a minimal proof-of-concept quasilinear calculation aiming to put the <property>shear dynamo</property>, a new effect recently found in numerical experiments, on a firm theoretical footing. Numerically observed scalings of the <property>wavenumber</property> and <property>growth rate</property> of the fastest growing <object>mode</object>, previously not understood, are derived analytically. The simplicity of the model suggests that <property>shear dynamo action</property> may be a generic property of sheared <object>magnetohydrodynamic turbulence</object>.", "tagged_phrases": {"task": ["Large-Scale Magnetic-Field Generation", "generation of a large-scale magnetic field"], "method": ["Randomly Forced Shearing Waves", "random non-helically forced motions", "linear shear"], "object": ["magnetic field", "conducting fluid", "electromotive force", "mode", "magnetohydrodynamic turbulence"], "property": ["dynamo", "kinematic", "shear dynamo", "wavenumber", "growth rate", "shear dynamo action"]}}
+{"title": "Sunyaev-Zel'dovich Effect Observations of the Bullet Cluster (1E 0657-56) with APEX-SZ", "abstract": "We present observations of the Sunyaev-Zel'dovich effect (SZE) in the Bullet cluster (1E 0657--56) using the APEX-SZ instrument at 150 GHz with a resolution of 1 arcmin. The main results are maps of the SZE in this massive, merging galaxy cluster. The cluster is detected with 23 sigma significance within the central 1 arcmin radius of the source position. The SZE map has a broadly similar morphology to that in existing X-ray maps of this system, and we find no evidence for significant contamination of the SZE emission by radio or IR sources. In order to make simple quantitative comparisons with cluster gas models derived from X-ray observations, we fit our data to an isothermal elliptical beta model, despite the inadequacy of such a model for this complex merging system. With an X-ray derived prior on the power-law index, beta = 1.04 +0.16 -0.10, we find a core radius r_c =142 +/- 18 arcsec, an axial ratio of 0.889 +/- 0.072, and a central temperature decrement of -771 +/- 71 micro-K_CMB, including a +/-5.5% flux calibration uncertainty. Combining the APEX-SZ map with a map of projected electron surface density from Chandra X-ray observations, we determine the mass-weighted temperature of the cluster gas to be T_mg=10.8 +/- 0.9 keV, significantly lower than some previously reported X-ray spectroscopic temperatures. Under the assumption of an isothermal cluster gas distribution in hydrostatic equilibrium, we compute the gas mass fraction for prolate and oblate spheroidal geometries and find it to be consistent with previous results from X-ray and weak lensing observations. This work is the first result from the APEX-SZ experiment, and represents the first reported scientific result from observations with a large array of multiplexed superconducting transition-edge sensor bolometers.", "tagged_title": "<object>Sunyaev-Zel'dovich Effect</object> Observations of the <object>Bullet Cluster (1E 0657-56)</object> with <instrument>APEX-SZ</instrument>", "tagged_abstract": "We present observations of the <object>Sunyaev-Zel'dovich effect (SZE)</object> in the  <object>Bullet cluster (1E 0657--56) </object> using the <instrument>APEX-SZ</instrument> instrument at 150 GHz with a resolution of 1 arcmin. The main results are maps of the <object>SZE</object> in this massive, merging <object>galaxy cluster</object>. The <object>cluster</object> is detected with 23 sigma significance within the central 1 arcmin <property>radius</property> of the source position. The <object>SZE map</object> has a broadly similar morphology to that in existing <object><modality>X-ray</modality> maps</object> of this system, and we find no evidence for significant contamination of the <object>SZE emission</object> by <modality>radio</modality> or <modality>IR</modality> sources. In order to make simple quantitative comparisons with <model><object>cluster gas</object> models</model> derived from <dataset><modality>X-ray</modality> observations</dataset>, we fit our data to an <method>isothermal elliptical beta model</method>, despite the inadequacy of such a model for this complex merging system. With an <modality>X-ray</modality> derived prior on the <object>power-law index</object>, beta = 1.04 +0.16 -0.10, we find a core <property>radius</property> r_c =142 +/- 18 arcsec, an <property>axial ratio</property> of 0.889 +/- 0.072, and a central <property>temperature decrement</property> of -771 +/- 71 micro-K_CMB, including a +/-5.5% <property>flux calibration uncertainty</property>. Combining the  <object><instrument>APEX-SZ</instrument> map </object> with a map of projected electron surface density from <dataset><instrument>Chandra<instrument> <modality>X-ray</modality> observations</dataset>, we determine the <property>mass-weighted</property> <property>temperature</property> of the <object>cluster gas</object> to be T_mg=10.8 +/- 0.9 keV, significantly lower than some previously reported <modality>X-ray spectroscopic</modality> temperatures. Under the assumption of an isothermal <object>cluster gas</object> distribution in hydrostatic equilibrium, we compute the <property>gas mass fraction</property> for <object>prolate</object> and <object>oblate spheroidal geometries</object> and find it to be consistent with previous results from <modality>X-ray</modality> and <modality>weak lensing</modality> observations. This work is the first result from the <instrument>APEX-SZ</instrument> experiment, and represents the first reported scientific result from observations with a large array of <instrument>multiplexed superconducting transition-edge sensor bolometers</instrument>.", "tagged_phrases": {"object": ["Sunyaev-Zel'dovich Effect", "Bullet Cluster (1E 0657-56)", "Sunyaev-Zel'dovich effect (SZE)", "Bullet cluster (1E 0657--56)", "SZE", "galaxy cluster", "cluster", "SZE map", "X-ray maps", "SZE emission", "cluster gas", "power-law index", "APEX-SZ map", "prolate", "oblate spheroidal geometries"], "instrument": ["APEX-SZ", "Chandra X-ray observations", "X-ray observations", "multiplexed superconducting transition-edge sensor bolometers"], "property": ["radius", "axial ratio", "temperature decrement", "flux calibration uncertainty", "mass-weighted", "temperature", "gas mass fraction"], "modality": ["X-ray", "radio", "IR", "X-ray spectroscopic", "weak lensing"], "model": ["cluster gas models"], "dataset": ["X-ray observations", "Chandra X-ray observations"], "method": ["isothermal elliptical beta model"]}}
+{"title": "Natural Coordinate System in Curved Space-time", "abstract": "In this paper we establish a generally and globally valid coordinate system in curved space-time with the simultaneous hypersurface orthogonal to the time coordinate. The time coordinate can be preseted according to practical evolving process and keep synchronous with the evolution of the realistic world. In this coordinate system, it is convenient to express the physical laws and to calculate physical variables with clear geometrical meaning. We call it \"natural coordinate system\". The constructing method for the natural coordinate system is concretely provided, and its physical and geometrical meanings are discussed in detail. In NCS we make classical approximation of spinor equation to get Newtonian mechanics, and then make weak field approximation of Einstein's equation and low speed approximation of particles moving in the space-time. From the analysis and examples we find it is a nice coordinate system to describe the realistic curved space-time, and is helpful to understand the nature of space-time.", "tagged_title": "<model>Natural Coordinate System</model> in <object>Curved Space-time</object>", "tagged_abstract": "In this paper we establish a generally and globally valid <model>coordinate system</model> in <object>curved space-time</object> with the simultaneous <object>hypersurface orthogonal</object> to the <object>time coordinate</object>. The <object>time coordinate</object> can be preseted according to practical evolving process and keep synchronous with the evolution of the realistic world. In this <model>coordinate system</model>, it is convenient to express the physical laws and to calculate physical variables with clear <property>geometrical</property> meaning. We call it <model>\"natural coordinate system\"</model>. The constructing method for the <model>natural coordinate system</model> is concretely provided, and its <property>physcial</property> and <property>geometrical</property> meanings are discussed in detail. In <model>NCS</model> we make <method>classical approximation of spinor equation</method> to get <object>Newtonian mechanics</object>, and then make <method>weak field approximation of Einstein's equation</method> and <method>low speed approximation of particles</method> moving in the <object>space-time</object>. From the analysis and examples we find it is a nice <model>coordinate system</model> to describe the realistic <object>curved space-time</object>, and is helpful to understand the nature of <object>space-time</object>.", "tagged_phrases": {"model": ["Natural Coordinate System", "coordinate system", "\"natural coordinate system\"", "natural coordinate system", "NCS"], "object": ["Curved Space-time", "curved space-time", "hypersurface orthogonal", "time coordinate", "Newtonian mechanics", "space-time"], "property": ["geometrical", "physcial"], "method": ["classical approximation of spinor equation", "weak field approximation of Einstein's equation", "low speed approximation of particles"]}}
+{"title": "2MTF I. The Tully-Fisher Relation in the 2MASS J, H and K Bands", "abstract": "The 2 Micron All-Sky Survey (2MASS) Tully-Fisher Survey (2MTF) aims to measure Tully-Fisher (TF) distances to all bright inclined spirals in the 2MASS Redshift Survey (2MRS). Essential to this project is a universal calibration of the TF relation in the 2MASS J (1.2 um), H (1.6 um) and K (2.2 um) bands. We present the first bias corrected or universal TF template in these bands. We find that the slope of the TF relation becomes steeper as the wavelength increases being close to L \\propto v^4 in K-band and L \\propto v^3.6 in J and H-bands. We also investigate the dependence on galaxy morphology showing that in all three bands the relation is steeper for later type spirals which also have a dimmer TF zeropoint than earlier type spirals. We correct the final relation to that for Sc galaxies. Finally we study the scatter from the TF relation fitting for a width dependent intrinsic scatter which is not found to vary significantly with wavelength.", "tagged_title": "2MTF I. <method>The Tully-Fisher Relation</method> in <object>the 2MASS J, H and K Bands</object>", "tagged_abstract": "<dataset>The 2 Micron All-Sky Survey (2MASS)</dataset> <dataset>Tully-Fisher Survey (2MTF)</dataset> aims to measure <property>Tully-Fisher (TF) distances</property> to all <object>bright inclined spirals</object> in <dataset>the 2MASS Redshift Survey (2MRS)</dataset>. Essential to this project is a <method>universal calibration</method> of the <method>TF relation</method> in <object>the 2MASS J (1.2 um), H (1.6 um) and K (2.2 um) bands</object>. We present the first <model>bias corrected or universal TF template</model> in these bands. We find that the slope of the <method>TF relation</method> becomes steeper as <property>the wavelength</property> increases being close to L \\propto v^4 in <object>K-band</object> and L \\propto v^3.6 in <object>J and H-bands</object>. We also investigate the dependence on <property>galaxy morphology</property> showing that in all three bands the relation is steeper for <object>later type spirals</object> which also have a dimmer <property>TF zeropoint</property> than <object>earlier type spirals</object>. We correct the final relation to that for <object>Sc galaxies</object>. Finally we study the scatter from the <method>TF relation</method> fitting for a width dependent intrinsic scatter which is not found to vary significantly with <object>wavelength</object>.", "tagged_phrases": {"method": ["The Tully-Fisher Relation", "universal calibration", "TF relation"], "object": ["the 2MASS J, H and K Bands", "bright inclined spirals", "the 2MASS J (1.2 um), H (1.6 um) and K (2.2 um) bands", "K-band", "J and H-bands", "later type spirals", "earlier type spirals", "Sc galaxies", "wavelength"], "dataset": ["The 2 Micron All-Sky Survey (2MASS)", "Tully-Fisher Survey (2MTF)", "the 2MASS Redshift Survey (2MRS)"], "property": ["Tully-Fisher (TF) distances", "the wavelength", "galaxy morphology", "TF zeropoint"], "model": ["bias corrected or universal TF template"]}}
+{"title": "The S1 Shell and Interstellar Magnetic Field and Gas near the Heliosphere", "abstract": "Many studies of the Loop I magnetic superbubble place the Sun at the edges of the bubble. One recent study models the polarized radio continuum of Loop I as two magnetic shells with the Sun embedded in the rim of the 'S1' shell. If the Sun is in such a shell, it should be apparent in both the local interstellar magnetic field and the distribution of nearby interstellar material. The properties of these subshells are compared to optical polarization data that trace the nearby interstellar magnetic field (ISMF), and the distribution of interstellar FeII and CaII within about 55 pc of the Sun. The result is that a model of Loop I as composed of two magnetic subshells is a viable description of the distribution of nearby low density ISM. Column densities of the recombinant ion CaII are found to increase with the strength of the interstellar radiation field, rather than with star distance or total pathlength through the two magnetic subshells. As found earlier, the symmetry of the dipole moment of the cosmic microwave background corresponds to the symmetries of the heliosphere and local ISMF (to within 10 degrees).", "tagged_title": "The <object>S1 Shell</object> and <object>Interstellar Magnetic Field</object> and Gas near <object>the Heliosphere</object>", "tagged_abstract": "Many studies of the <object>Loop I magnetic superbubble</object> place <object>the Sun</object> at the edges of the bubble. One recent study models <object>the polarized radio continuum</object> of <object>Loop I</object> as two <object>magnetic shells</object> with <object>the Sun</object> embedded in the rim of <object>the 'S1' shell</object>. If <object>the Sun</object> is in such a shell, it should be apparent in both <object>the local interstellar magnetic field</object> and the distribution of nearby <object>interstellar material</object>. The properties of these subshells are compared to <modality>optical polarization data</modality> that trace the nearby <object>interstellar magnetic field (ISMF)</object>, and the distribution of <object>interstellar FeII and CaII</object> within about 55 pc of <object>the Sun</object>. The result is that a model of <object>Loop I</object> as composed of two <object>magnetic subshells</object> is a viable description of the distribution of nearby <object>low density ISM</object>. <property>Column densities</property> of <object>the recombinant ion CaII</object> are found to increase with the strength of <object>the interstellar radiation field</object>, rather than with <property>star distance</property> or <property>total pathlength</property> through the two <object>magnetic subshells</object>. As found earlier, the symmetry of <property>the dipole moment</property> of <object>the cosmic microwave background</object> corresponds to the symmetries of <object>the heliosphere</object> and <object>local ISMF</object> (to within 10 degrees).", "tagged_phrases": {"object": ["S1 Shell", "Interstellar Magnetic Field", "the Heliosphere", "Loop I magnetic superbubble", "the Sun", "the polarized radio continuum", "Loop I", "magnetic shells", "the 'S1' shell", "the local interstellar magnetic field", "interstellar material", "interstellar magnetic field (ISMF)", "interstellar FeII and CaII", "magnetic subshells", "low density ISM", "the recombinant ion CaII", "the interstellar radiation field", "the cosmic microwave background", "the heliosphere", "local ISMF"], "modality": ["optical polarization data"], "property": ["Column densities", "star distance", "total pathlength", "the dipole moment"]}}
+{"title": "The growth of supermassive black holes in pseudo-bulges, classical bulges and elliptical galaxies", "abstract": "Using results from structural analysis of a sample of nearly 1000 local galaxies from the Sloan Digital Sky Survey, we estimate how the mass in central black holes is distributed amongst elliptical galaxies, classical bulges and pseudo-bulges, and investigate the relation between their stellar masses and central stellar velocity dispersion sigma. Assuming a single relation between elliptical galaxy/bulge mass, M_Bulge, and central black hole mass, M_BH, we find that 55^{+8}_{-4} per cent of the mass in black holes in the local universe is in the centres of elliptical galaxies, 41^{+4}_{-2} per cent in classical bulges and 4^{+0.9}_{-0.4} per cent in pseudo-bulges. We find that ellipticals, classical bulges and pseudo-bulges follow different relations between their stellar masses and sigma, and the most significant offset occurs for pseudo-bulges in barred galaxies. This structural dissimilarity leads to discrepant black hole masses if single M_BH-M_Bulge and M_BH-sigma relations are used. Adopting relations from the literature, we find that the M_BH-sigma relation yields an estimate of the total mass density in black holes that is roughly 55 per cent larger than if the M_BH-M_Bulge relation is used.", "tagged_title": "The growth of <object>supermassive black holes</object> in <object>pseudo-bulges</object>, <object>classical bulges</object> and <object>elliptical galaxies</object>", "tagged_abstract": "Using results from <method>structural analysis</method> of a sample of nearly 1000 <object>local galaxies</object> from the <dataset>Sloan Digital Sky Survey</dataset>, we estimate how the <property>mass</property> in <object>central black holes</object> is distributed amongst <object>elliptical galaxies</object>, <object>classical bulges</object> and <object>pseudo-bulges</object>, and investigate the relation between their <property>stellar masses</property> and <property>central stellar velocity dispersion sigma</property>. Assuming a single relation between <property>elliptical galaxy/bulge mass</property>, M_Bulge, and <property>central black hole mass</property>, M_BH, we find that 55^{+8}_{-4} per cent of the <property>mass</property> in <object>black holes</object> in the <object>local universe</object> is in the centres of <object>elliptical galaxies</object>, 41^{+4}_{-2} per cent in <object>classical bulges</object> and 4^{+0.9}_{-0.4} per cent in <object>pseudo-bulges</object>. We find that <object>ellipticals</object>, <object>classical bulges</object> and <object>pseudo-bulges</object> follow different relations between their <property>stellar masses</property> and <property>sigma</property>, and the most significant offset occurs for <object>pseudo-bulges</object> in <object>barred galaxies</object>. This structural dissimilarity leads to discrepant <property>black hole masses</property> if single M_BH-M_Bulge and M_BH-sigma relations are used. Adopting relations from the literature, we find that the M_BH-sigma relation yields an estimate of the <property>total mass density</property> in <object>black holes</object> that is roughly 55 per cent larger than if the M_BH-M_Bulge relation is used.", "tagged_phrases": {"object": ["supermassive black holes", "pseudo-bulges", "classical bulges", "elliptical galaxies", "local galaxies", "central black holes", "black holes", "local universe", "ellipticals", "barred galaxies"], "method": ["structural analysis"], "dataset": ["Sloan Digital Sky Survey"], "property": ["mass", "stellar masses", "central stellar velocity dispersion sigma", "elliptical galaxy/bulge mass", "central black hole mass", "sigma", "black hole masses", "total mass density"]}}
+{"title": "Characteristic Scales of Initial Density and Velocity Fields", "abstract": "For the initial fields of the density contrast and peculiar velocity, we theoretically calculate the differential and integral length scales, i.e., statistical measures that respectively characterize the small- and large-scale fluctuations of a random field. These length scales and the associated mass scales explain the length and mass scales observed for (1) halos of young galaxies at z > 5, (2) halos of galaxies at z = 0, and (3) the largest structures in the galaxy distribution at z = 0. We thereby discuss that such observed scales are fossil imprints of the characteristic scales of the initial fields.", "tagged_title": "<property>Characteristic Scales</property> of <object>Initial Density and Velocity Fields</object>", "tagged_abstract": "For the <object>initial fields</object> of the <object>density contrast</object> and <object>peculiar velocity</object>, we theoretically calculate the <property>differential and integral length scales</property>, i.e., statistical measures that respectively characterize the small- and large-scale fluctuations of a <object>random field</object>. These <property>length scales</property> and the associated <property>mass scales</property> explain the <property>length and mass scales</property> observed for (1) <object>halos of young galaxies</object> at z > 5, (2) <object>halos of galaxies</object> at z = 0, and (3) the largest structures in the <object>galaxy distribution</object> at z = 0. We thereby discuss that such observed scales are fossil imprints of the <property>characteristic scales</property> of the <property>initial fields</property>.", "tagged_phrases": {"property": ["Characteristic Scales", "differential and integral length scales", "length scales", "mass scales", "length and mass scales", "characteristic scales", "initial fields"], "object": ["Initial Density and Velocity Fields", "initial fields", "density contrast", "peculiar velocity", "random field", "halos of young galaxies", "halos of galaxies", "galaxy distribution"]}}
+{"title": "Gravitino Dark Matter and the Flavour Structure of R-violating Operators", "abstract": "We study gravitino dark matter and slow gravitino decays within the framework of R-violating supersymmetry, with particular emphasis on the flavour dependence of the branching ratios and the allowed R-violating couplings. The dominant decay modes and final state products turn out to be very sensitive to the R-violating hierarchies. Mixing effects can be crucial in correctly deriving the relative magnitude of the various contributions, particularly for heavy flavours with phase space suppression. The study of the strength of different decay rates for the gravitino is also correlated to collider signatures expected from decays of the Next-to-Lightest Supersymmetric Particle (NLSP) and to single superparticle production.", "tagged_title": "<object>Gravitino Dark Matter</object> and the Flavour Structure of <object>R-violating Operators</object>", "tagged_abstract": "We study <object>gravitino dark matter</object> and slow <object>gravitino decays</object> within the framework of <object>R-violating supersymmetry</object>, with particular emphasis on the <property>flavour dependence</property> of the branching ratios and the allowed <object>R-violating couplings</object>. The dominant <property>decay modes</property> and <property>final state products</property> turn out to be very sensitive to the R-violating hierarchies. <method>Mixing effects</method> can be crucial in correctly deriving the relative <property>magnitude</property> of the various contributions, particularly for heavy flavours with <property>phase space suppression</property>. The study of the <property>strength of different decay rates</property> for the <object>gravitino</object> is also correlated to collider signatures expected from decays of the <object>Next-to-Lightest Supersymmetric Particle</object> (NLSP) and to single <object>superparticle</object> production.", "tagged_phrases": {"object": ["Gravitino Dark Matter", "R-violating Operators", "gravitino dark matter", "gravitino decays", "R-violating supersymmetry", "R-violating couplings", "gravitino", "Next-to-Lightest Supersymmetric Particle", "superparticle"], "property": ["flavour dependence", "decay modes", "final state products", "magnitude", "phase space suppression", "strength of different decay rates"], "method": ["Mixing effects"]}}
+{"title": "Dark Matter Caustics in Galaxy Clusters", "abstract": "We interpret the recent gravitational lensing observations of Jee et al. \\cite{Jee} as first evidence for a {\\it caustic} ring of dark matter in a galaxy cluster. A caustic ring unavoidably forms when a cold collisionless flow falls with net overall rotation in and out of a gravitational potential well. Evidence for caustic rings of dark matter was previously found in the Milky Way and other isolated spiral galaxies. We argue that galaxy clusters have at least one and possibly two or three caustic rings. We calculate the column density profile of a caustic ring in a cluster and show that it is consistent with the observations of Jee et al.", "tagged_title": "<object>Dark Matter Caustics</object> in <object>Galaxy Clusters</object>", "tagged_abstract": "We interpret the recent <modality><object>gravitational lensing</object> observations</modality> of Jee et al. \\cite{Jee} as first evidence for a <object>{\\it caustic} ring</object> of <object>dark matter</object> in a <object>galaxy cluster</object>. A <object>caustic ring</object> unavoidably forms when a <property>cold</property> <property>collisionless</property> <object>flow</object> falls with <property>net overall rotation</property> in and out of a <object>gravitational potential well</object>. Evidence for <object>caustic rings</object> of <object>dark matter</object> was previously found in the <object>Milky Way</object> and other <property>isolated</property> <property>spiral</property> <object>galaxies</object>. We argue that <object>galaxy clusters</object> have at least one and possibly two or three <object>caustic rings</object>. We <task>calculate the <property>column density profile</property></task> of a <object>caustic ring</object> in a <object>cluster</object> and <task>show that it is consistent with the observations of</task> Jee et al.", "tagged_phrases": {"object": ["Dark Matter Caustics", "Galaxy Clusters", "gravitational lensing", "caustic ring", "dark matter", "galaxy cluster", "flow", "gravitational potential well", "caustic rings", "Milky Way", "galaxies", "galaxy clusters", "cluster"], "modality": ["gravitational lensing observations"], "property": ["cold", "collisionless", "net overall rotation", "isolated", "spiral", "column density profile"], "task": ["calculate the column density profile", "show that it is consistent with the observations of"]}}
+{"title": "Dark matter in elliptical galaxies: II. Estimating the mass within the virial radius", "abstract": "Elliptical galaxies are modelled with a a 4-component model: Sersic stars, LCDM dark matter (DM), hot gas and central black hole. DM is negligible in the inner regions, which are dominated by stars and the central black hole. This prevents any kinematical estimate (using a Jeans analysis) of the inner slope of the DM density profile. The gas fraction rises, but the baryon fraction decreases with radius, at least out to 10 effective radii (R_e). Even with line-of-sight velocity dispersion (VD) measurements at 4 to 6 R_e with 20 km/s accuracy and perfectly known velocity anisotropy, the total mass within the virial radius (r_v) is uncertain by a factor over 3. The DM distributions found in LCDM simulations are consistent with the stellar VD profiles, but appear inconsistent with the low VDs measured by Romanowsky et al. (2003) of planetary nebulae between 2 and 5 R_e, which imply such low M/Ls that the baryon fraction within r_v must be greater than the universal value. Replacing the NFW DM model by the new model of Navarro et al. (2004) decreases slightly the VD at a given radius. So, given the observed VD measured at 5 R_e, the inferred M/L within r_v is 40% larger than predicted with the NFW model. Folding in the slight (strong) radial anisotropy found in LCDM (merger) simulations, which is well modelled (much better than with the Osipkov-Merritt formula) with beta(r) = 1/2 r/(r+a), the inferred M/L within r_v is another 1.6 (2.4) times higher than for the isotropic NFW model. Thus, the DM model and radial anisotropy can partly explain the low PN VDs, but not in full. In an appendix, single integral expressions are derived for the VDs in terms of the tracer density and total mass profiles, for 3 anisotropic models: radial, Osipkov-Merritt, and the model above, for general radial profiles of luminosity density and mass.", "tagged_title": "<object>Dark matter</object> in <object>elliptical galaxies</object>: II. <task>Estimating the <property>mass</property> within the <property>virial radius</property></task>", "tagged_abstract": "<object>Elliptical galaxies</object> are <task>modelled</task> with a <model><property>4-component</property> model</model>: <object>Sersic stars</object>, <model>LCDM</model> <object>dark matter</object> (DM), <property>hot</property> <object>gas</object> and <property>central</property> <object>black hole</object>. DM is negligible in the <object>inner regions</object>, which are dominated by <object>stars</object> and the <property>central</property> <object>black hole</object>. This prevents any kinematical estimate (using a <method>Jeans analysis</method>) of the <property>inner slope</property> of the <object>DM density profile</object>. The <property>gas fraction</property> rises, but the <property>baryon fraction</property> decreases with <property>radius</property>, at least out to 10 <property>effective radii</property> (R_e). Even with <property>line-of-sight velocity dispersion</property> (VD) measurements at 4 to 6 R_e with 20 km/s accuracy and perfectly known <property>velocity anisotropy</property>, the total <property>mass</property> within the <property>virial radius</property> (r_v) is uncertain by a factor over 3. The DM distributions found in <method><model>LCDM</model> simulations</method> are consistent with the <modality>stellar VD profiles</modality>, but appear inconsistent with the low VDs measured by Romanowsky et al. (2003) of <object>planetary nebulae</object> between 2 and 5 R_e, which imply such low M/Ls that the <property>baryon fraction</property> within r_v must be greater than the universal value. Replacing the <model>NFW DM model</model> by the new <model>model of Navarro et al. (2004)</model> decreases slightly the VD at a given <property>radius</property>. So, given the observed VD measured at 5 R_e, the inferred M/L within r_v is 40% larger than predicted with the <model>NFW model</model>. Folding in the slight (strong) <property>radial anisotropy</property> found in <method><model>LCDM</model> (merger) simulations</method>, which is well modelled (much better than with the <model>Osipkov-Merritt formula</model>) with beta(r) = 1/2 r/(r+a), the inferred M/L within r_v is another 1.6 (2.4) times higher than for the <model>isotropic NFW model</model>. Thus, the <model>DM model</model> and <property>radial anisotropy</property> can partly explain the low PN VDs, but not in full. In an appendix, <task><method>single integral expressions</method> are derived</task> for the VDs in terms of the <property>tracer density</property> and <property>total mass</property> profiles, for 3 <model>anisotropic models</model>: <model>radial</model>, <model>Osipkov-Merritt</model>, and the model above, for general <property>radial profiles</property> of <property>luminosity</property> <property>density</property> and <property>mass</property>.", "tagged_phrases": {"object": ["Dark matter", "elliptical galaxies", "Elliptical galaxies", "Sersic stars", "dark matter", "gas", "black hole", "inner regions", "stars", "DM density profile", "planetary nebulae"], "task": ["Estimating the mass within the virial radius", "modelled", "single integral expressions are derived"], "property": ["mass", "virial radius", "4-component", "hot", "central", "inner slope", "gas fraction", "baryon fraction", "radius", "effective radii", "line-of-sight velocity dispersion", "velocity anisotropy", "radial anisotropy", "tracer density", "total mass", "radial profiles", "luminosity", "density"], "model": ["4-component model", "LCDM", "NFW DM model", "model of Navarro et al. (2004)", "NFW model", "Osipkov-Merritt formula", "isotropic NFW model", "DM model", "anisotropic models", "radial", "Osipkov-Merritt"], "method": ["Jeans analysis", "LCDM simulations", "LCDM (merger) simulations", "single integral expressions"], "modality": ["stellar VD profiles"]}}
+{"title": "On the appearance of hyperons in neutron stars", "abstract": "By employing a recently constructed hyperon-nucleon potential the equation of state of \\beta-equilibrated and charge neutral nucleonic matter is calculated. The hyperon-nucleon potential is a low-momentum potential which is obtained within a renormalization group framework. Based on the Hartree-Fock approximation at zero temperature the densities at which hyperons appear in neutron stars are estimated. For several different bare hyperon-nucleon potentials and a wide range of nuclear matter parameters it is found that hyperons in neutron stars are always present. These findings have profound consequences for the mass and radius of neutron stars.", "tagged_title": "On the appearance of <object>hyperons</object> in <object>neutron stars</object>", "tagged_abstract": "By employing a recently constructed <model>hyperon-nucleon potential</model> the <property>equation of state</property> of <object>\\beta-equilibrated and charge neutral nucleonic matter</object> is calculated. The <model>hyperon-nucleon potential</model> is a low-<property>momentum</property> potential which is obtained within a <model>renormalization group</model> framework. Based on the <model>Hartree-Fock approximation</model> at zero <property>temperature</property> <task>the <property>densities</property> at which <object>hyperons</object> appear in <object>neutron stars</object> are estimated</task>. For several different <model>bare hyperon-nucleon potentials</model> and a wide range of nuclear matter parameters it is found that <object>hyperons</object> in <object>neutron stars</object> are always present. These findings have profound consequences for the <property>mass</property> and <property>radius</property> of <object>neutron stars</object>.", "tagged_phrases": {"object": ["hyperons", "neutron stars", "\\beta-equilibrated and charge neutral nucleonic matter"], "model": ["hyperon-nucleon potential", "renormalization group", "Hartree-Fock approximation", "bare hyperon-nucleon potentials"], "property": ["equation of state", "momentum", "temperature", "densities", "mass", "radius"], "task": ["the densities at which hyperons appear in neutron stars are estimated"]}}

data/manual/schema.json

@@ -0,0 +1,11 @@
+{
+    "model": "a representation of a (scientific) phenomenon using mathematical formalism and/or computational simulation",
+    "task": "a specific problem, objective or goal to be accomplished",
+    "dataset": "a collection of data, measurements or observations",
+    "field": "an academic (sub)discipline",
+    "modality": "a class or type of data/observations with similar or the same structure",
+    "method": "an approach, technique or procedure to complete a task",
+    "object": "an entity that can be studied",
+    "property": "a quantitative or qualitative descriptor, or an inherent attribute of an entity, data, modality or method",
+    "instrument": "a device or system used for making measurements"
+}

data/raw/README.md

@@ -0,0 +1,13 @@
+```
+raw
+├── astro_physpace
+├── evo_bio
+├── fluid_dyn
+├── mmu
+├── phy_strat
+└── the_well
+```
+-----------------------
+- The `raw` folder contains the arXiv scraped data for different arxiv categories.
+- The data is stored in manually named folders, named after the field of research, eg. fluid dynamics, evolutionary biology, astrophysics
+- Each file inside the folders in `raw` contains jsonlines file named after the arxiv category, the number of papers scraped for that channel from arxiv, the date it was scraped and the order of the papers that were scraped.

data/raw/astro_physpace/astro-ph_6000_last_updated_date_desc_20240715_134752.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b004b1bd0385e3b768e5d28f4f1d3742cb8f3d495ef21cc6abbf1ae09e03562
+size 10592937

data/raw/evo_bio/q-bio.PE_10000_last_updated_date_desc_20240723_163228.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76f9766cc5c5c9298a15a78497011a8f1f88b5c06df6f88339e88b8694b021a9
+size 18023838

data/raw/fluid_dyn/physics.flu-dyn_10000_last_updated_date_desc_20240723_155932.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c3955f85bc3f6c6fd1d098a6087c499ae0851fc205657d32920aeb5e9563eefd
+size 19089885

data/raw/phy_strat/stratified_arxiv_data_combined_astro_physpace.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e26d95c6a991254d0c3beab62db2171b19490491998094eb42d778dfc87eeeb4
+size 10640422

data/raw/the_well/the_well_papers.jsonl

@@ -0,0 +1,21 @@
+{"dataset": "MHD_256", "url": "https://iopscience.iop.org/article/10.3847/1538-4357/abc484/pdf", "links": "https://iopscience.iop.org/article/10.3847/1538-4357/abc484/pdf", "abstract": "An essential component of the solar wind, galaxy formation, and of interstellar medium (ISM) dynamics is magnetohydrodynamic (MHD) turbulence. This dataset consists of isothermal MHD simulations without self-gravity (such as found in the diffuse ISM) initially generated with resolution $256^3$ and then downsampled to $64^3$ after anti-aliasing with an ideal low-pass filter.\nDimension of discretized data: 100 timesteps of $256\\times 256\\times256$ cubes.\nFields available in the data: Density (scalar field), velocity (vector field), magnetic field (vector field).\nNumber of trajectories: 10 Initial conditions x 10 combination of parameters = 100 trajectories.\nEstimated size of the ensemble of all simulations: 4.58TB.\nGrid type: uniform grid, cartesian coordinates.\nInitial conditions: uniform IC.\nBoundary conditions: periodic boundary conditions.\nData are stored separated by ($\\Delta t$): 0.01 (arbitrary units).\nTotal time range ($t\\_{min}$ to $t\\_{max}$): $t\\_{min} = 0$, $t\\_{max} = 1$.\nSpatial domain size ($L_x$, $L_y$, $L_z$): dimensionless so 256 pixels.\nSet of coefficients or non-dimensional parameters evaluated: all combinations of $\\mathcal{M}_s=${0.5, 0.7, 1.5, 2.0 7.0} and $\\mathcal{M}_A =${0.7, 2.0}.\nApproximate time to generate the data: 48 hours per simulation.\nHardware used to generate the data**: 64 cores.\nWhat phenomena of physical interest are catpured in the data: MHD fluid flows in the compressible limit (sub and super sonic, sub and super Alfvenic).\nHow to evaluate a new simulator operating in this space: Check metrics such as Power spectrum, two points correlation function.", "id": "Magnetohydrodynamics (MHD) compressible turbulence", "primary_category": "physics.flu-dyn", "authors": "['B. Burkhart',  'S. M. Appel',  'S. Bialy',  'J. Cho',  'A. J. Christensen',  'D. Collins',  'C. Federrath',  'D. B. Fielding',  'D. Finkbeiner',  'A. S. Hill',  'J. C. Ibáñez-Mejía',  'M. R. Krumholz',  'A. Lazarian',  'M. Li',  'P. Mocz',  'M.-M. Mac Low',  'J. Naiman',  'S. K. N. Portillo',  'B. Shane',  'Z. Slepian',  'Y. Yuan']", "updated": null}
+{"dataset": "MHD_64", "url": "https://iopscience.iop.org/article/10.3847/1538-4357/abc484/pdf", "links": "https://iopscience.iop.org/article/10.3847/1538-4357/abc484/pdf", "abstract": "An essential component of the solar wind, galaxy formation, and of interstellar medium (ISM) dynamics is magnetohydrodynamic (MHD) turbulence. This dataset consists of isothermal MHD simulations without self-gravity (such as found in the diffuse ISM) initially generated with resolution $256^3$ and then downsampled to $64^3$ after anti-aliasing with an ideal low-pass filter. This dataset is the downsampled version.\nDimension of discretized data: 100 timesteps of $64\\times 64\\times 64$ cubes.\nFields available in the data: Density (scalar field), velocity (vector field), magnetic field (vector field).\nNumber of trajectories: 10 Initial conditions x 10 combination of parameters = 100 trajectories.\nEstimated size of the ensemble of all simulations: 71.6 GB.\nGrid type: uniform grid, cartesian coordinates.\nInitial conditions: uniform IC.\nBoundary conditions: periodic boundary conditions.\nData are stored separated by ($\\Delta t$): 0.01 (arbitrary units).\nTotal time range ($t\\_{min}$ to $t\\_{max}$): $t\\_{min} = 0$, $t\\_{max} = 1$.\nSpatial domain size ($L_x$, $L_y$, $L_z$): dimensionless so 64 pixels.\nSet of coefficients or non-dimensional parameters evaluated: all combinations of $\\mathcal{M}_s=${0.5, 0.7, 1.5, 2.0 7.0} and $\\mathcal{M}_A =${0.7, 2.0}.\nApproximate time to generate the data: Downsampled from MHD_256 after applying ideal low-pass filter.\nHardware used to generate the data**: Downsampled from MHD_256 after applying ideal low-pass filter.\nWhat phenomena of physical interest are catpured in the data: MHD fluid flows in the compressible limit (sub and super sonic, sub and super Alfvenic).\nHow to evaluate a new simulator operating in this space: Check metrics such as Power spectrum, two-points correlation function.", "id": "Magnetohydrodynamics (MHD) compressible turbulence", "primary_category": "physics.flu-dyn", "authors": "['B. Burkhart',  'S. M. Appel',  'S. Bialy',  'J. Cho',  'A. J. Christensen',  'D. Collins',  'C. Federrath',  'D. B. Fielding',  'D. Finkbeiner',  'A. S. Hill',  'J. C. Ibáñez-Mejía',  'M. R. Krumholz',  'A. Lazarian',  'M. Li',  'P. Mocz',  'M.-M. Mac Low',  'J. Naiman',  'S. K. N. Portillo',  'B. Shane',  'Z. Slepian',  'Y. Yuan']", "updated": null}
+{"dataset": "acoustic_scattering_discontinuous_2d", "url": null, "links": null, "abstract": "These variable-coefficient acoustic equations describe the propogation of an acoustic pressure wave through domains consisting of multiple materials with different scattering properties. This problem emerges in source optimization and it's inverse - that of identifying the material properties from the scattering of the wave - is a vital problem in geology and radar design. This is the simplest of three scenarios. In this case, we have a variable number of initial point sources and single discontinuity separating two sub-domains. Within each subdomain, the density of the underlying material varies smoothly.\nDimension of discretized data: $101$ steps of $256\\times 256$ images.\nFields available in the data: pressure (scalar field), material density (constant scalar field), material speed of sound (constant scalar field), velocity field (vector field). \nNumber of trajectories: 2000.\nEstimated size of the ensemble of all simulations: 157.7 GB.\nGrid type: uniform, cartesian coordinates.\nInitial conditions: Flat pressure static field with 1-4 high pressure rings randomly placed in domain. The rings are defined with variable intensity $\\sim \\mathcal U(.5, 2)$ and radius $\\sim \\mathcal U(.06, .15)$. \nBoundary conditions: Open domain in $y$, reflective walls in $x$.\nSimulation time-step: Variable based on CFL with safety factor .25. \nData are stored separated by ($\\Delta t$): 2/101. \nTotal time range ($t_{min}$ to $t_{max}$): [0, 2.]\nSpatial domain size ($L_x$, $L_y$, $L_z$): [-1, 1] x [-1, 1]\nSet of coefficients or non-dimensional parameters evaluated:\n$K$ is fixed at 4.0. \n$\rho$ is the primary coefficient here. Each side is generated with one of the following distributions:\n- Gaussian Bump - Peak density samples from $\\sim\\mathcal U(1, 7)$ and $\\sigma \\sim\\mathcal U(.1, 5)$ with the center of the bump uniformly sampled from the extent of the subdomain.\n- Linear gradient - Four corners sampled with $\rho \\sim \\mathcal U(1, 7)$. Inner density is bilinearly interpolated.\n- Constant - Constant $\rho \\sim\\mathcal U(1, 7)$\n- Smoothed Gaussian Noise - Constant background sampled $\rho \\sim\\mathcal U(1, 7)$ with IID standard normal noise applied. This is then smoothed by a Gaussian filter of varying sigma $\\sigma \\sim\\mathcal U(5, 10)$\nApproximate time to generate the data: ~15 minutes per simulation. \nHardware used to generate the data and precision used for generating the data: 64 Intel Icelake cores per simulation. Generated in double precision.\nWave propogation through discontinuous media. Most existing machine learning datasets for computational physics are highly smooth and the acoustic challenges presented here offer challenging discontinuous scenarios that approximate complicated geometry through the variable density.", "id": "Acoustic Scattering - Single Discontinuity", "primary_category": "physics.flu-dyn", "authors": ["The Well Collaboration"], "updated": null}
+{"dataset": "acoustic_scattering_inclusions_2d", "url": null, "links": null, "abstract": "These variable-coefficient acoustic equations describe the propogation of an acoustic pressure wave through domains consisting of multiple materials with different scattering properties. This problem emerges in source optimization and it's inverse - that of identifying the material properties from the scattering of the wave - is a vital problem in geology and radar design. In this case, we have a variable number of initial point sources and a domain with random inclusions. These types of problems are of particular interest in geology where the inverse scattering is used to identify mineral deposits.\nDimension of discretized data: $101$ steps of $256\\times256$ images.\nFields available in the data: pressure (scalar field), material density (constant scalar field), material speed of sound (constant scalar field), velocity field (vector field).\nNumber of trajectories: 4000.\nEstimated size of the ensemble of all simulations: 283.8 GB.\nGrid type: uniform, cartesian coordinates.\nInitial conditions: Flat pressure static field with 1-4 high pressure rings randomly placed in domain. The rings are defined with variable intensity $\\sim \\mathcal U(.5, 2)$ and radius $\\sim \\mathcal U(.06, .15)$. \nBoundary conditions: Open domain in $y$, reflective walls in $x$.\nSimulation time-step: Variable based on CFL with safety factor .25. \nData are stored separated by ($\\Delta t$): 2/101. \nTotal time range ($t_{min}$ to $t_{max}$): [0, 2.].\nSpatial domain size ($L_x$, $L_y$, $L_z$): [-1, 1] x [-1, 1].\nSet of coefficients or non-dimensional parameters evaluated:\n$K$ is fixed at 4.0. \n$\rho$ is the primary coefficient here. This is a superset of the single discontinuity example so the background is first generated two splits with one of the following distributions:\n- Gaussian Bump - Peak density samples from $\\sim\\mathcal U(1, 7)$ and $\\sigma \\sim\\mathcal U(.1, 5)$ with the center of the bump uniformly sampled from the extent of the subdomain.\n- Linear gradient - Four corners sampled with $\rho \\sim \\mathcal U(1, 7)$. Inner density is bilinearly interpolated.\n- Constant - Constant $\rho \\sim\\mathcal U(1, 7)$\n- Smoothed Gaussian Noise - Constant background sampled $\rho \\sim\\mathcal U(1, 7)$ with IID standard normal noise applied. This is then smoothed by a Gaussian filter of varying sigma $\\sigma \\sim\\mathcal U(5, 10)$. \nInclusions are then added as 1-15 random ellipsoids with center uniformly sampled from the domain and height/width sampled uniformly from [.05, .6]. The ellipsoid is then rotated randomly with angle sampled [-45, 45]. For the inclusions, $Ln(\rho)\\sim \\mathcal U(-1, 10)$ \nApproximate time to generate the data: ~15 minutes per simulation. \nHardware used to generate the data and precision used for generating the data: 64 Intel Icelake cores per simulation. Generated in double precision.\nWave propogation through discontinuous media. Most existing machine learning datasets for computational physics are highly smooth and the acoustic challenges presented here offer challenging discontinuous scenarios that approximate complicated geometry through the variable density. The inclusions change wave propogation speed but only in small, irregular areas.", "id": "Acoustic Scattering - Inclusions", "primary_category": "physics.flu-dyn", "authors": ["The Well Collaboration"], "updated": null}
+{"dataset": "acoustic_scattering_maze_2d", "url": null, "links": null, "abstract": "These variable-coefficient acoustic equations describe the propogation of an acoustic pressure wave through maze-like domains. Pressure waves emerge from point sources and propogate through domains consisting of low density maze paths and orders of magnitude higher density maze walls. This is built primarily as a challenge for machine learning methods, though has similar properties to optimal placement problems like WiFi in a building.\nDimension of discretized data: $201$ steps of $256\\times256$ images.\nFields available in the data: pressure (scalar field), material density (constant scalar field), material speed of sound (constant scalar field), velocity field (vector field).\nNumber of trajectories: 2000.\nEstimated size of the ensemble of all simulations: 311.3 GB.\nGrid type: uniform, cartesian coordinates.\nInitial conditions: Flat pressure static field with 1-6 high pressure rings randomly placed along paths of maze. The rings are defined with variable intensity $\\sim \\mathcal U(3., 5.)$ and radius $\\sim \\mathcal U(.01, .04)$. Any overlap with walls is removed. \nBoundary conditions: Open domain in $y$, reflective walls in $x$.\nSimulation time-step: Variable based on CFL with safety factor .25. \nData are stored separated by ($\\Delta t$): 2/201. \nTotal time range ($t_{min}$ to $t_{max}$): [0,4.].\nSpatial domain size ($L_x$, $L_y$, $L_z$): [-1, 1] x [-1, 1].\nSet of coefficients or non-dimensional parameters evaluated:\n$K$ is fixed at 4.0. \n$\rho$ is the primary coefficient here. We generated a maze with initial width between 6 and 16 pixels and upsample it via nearest neighbor resampling to create a 256 x 256 maze. The walls are set to $\rho=10^6$ while paths are set to  $\rho=3$.  \nApproximate time to generate the data: ~20 minutes per simulation. \nHardware used to generate the data and precision used for generating the data: 64 Intel Icelake cores per simulation. Generated in double precision.\nThis is an example of simple dynamics in complicated geometry. The sharp discontinuities can be a significant problem for machine learning models, yet they are a common feature in many real-world physics. While visually the walls appear to stop the signal, it is actually simply the case that the speed of sound is much much lower inside the walls leading to partial reflection/absorbtion at the interfaces.", "id": "Acoustic Scattering - Maze", "primary_category": "physics.flu-dyn", "authors": ["The Well Collaboration"], "updated": null}
+{"dataset": "active_matter", "url": null, "links": null, "abstract": "Simulation of a continuum theory describing the dynamics of $N$ rod-like active particles immersed in a Stokes fluid having linear dimension $L$ and colume $L^2$.\nDimension of discretized data: $81$ time-steps of $256\\times256$ images per trajectory.\nFields available in the data: concentration (scalar field),\nvelocity (vector field), orientation tensor (tensor field), strain-rate tensor (tensor field).\nNumber of trajectories: $5$ trajectories per parameter-set, each trajectory being generated with a different initialization of the state field {$c,D,U$}.\nSize of the ensemble of all simulations: 51.3 GB.\nGrid type: Uniform grid, cartesian coordinates.\nInitial conditions: The concentration is set to constant value $c(x,t)=1$ and the orientation tensor is initialized as plane-wave perturbation about the isotropic state.\nBoundary conditions: Periodic boundary conditions.\nSimulation time-step: $3.90625\\times 10^{-4}$ seconds.\nData are stored separated by ($\\Delta t$): 0.25 seconds.\nTotal time range ($t_{min}$ to $t_{max}$): $0$ to $20$ seconds.\nSpatial domain size ($L_x$, $L_y$, $L_z$): $L_x=10$ and $L_y=10$\nSet of coefficients or non-dimensional parameters evaluated: $\u0007lpha =$ {-1,-2,-3,-4,-5}; $\beta  =$ {0.8}; \n$\\zeta =$ {1,3,5,7,9,11,13,15,17}.\nApproximate time and hardware to generate the data: 20 minutes per simulation on an A100 GPU in double precision. There is a total of 225 simulations, which is approximately 75 hours.\nWhat phenomena of physical interest are catpured in the data: How is energy being transferred between scales? How is vorticity coupled to the orientation field? Where does the transition from isotropic state to nematic state occur with the change in alignment ($\\zeta$) or dipole strength ($\u0007lpha$)? \nHow to evaluate a new simulator operating in this space: Reproducing some summary statistics like power spectra and average scalar order parameters. Additionally, being able to accurately capture the phase transition from isotropic to nematic state.", "id": "Active fluid simulations", "primary_category": "physics.flu-dyn", "authors": ["The Well Collaboration"], "updated": null}
+{"dataset": "convective_envelope_rsg", "url": "https://iopscience.iop.org/article/10.3847/1538-4357/ac5ab3", "links": "https://iopscience.iop.org/article/10.3847/1538-4357/ac5ab3", "abstract": "Massive stars evolve into red supergiants, which have large radii and luminosities, and low-density, turbulent, convective envelopes. These simulations model the (inherently 3D) convective properties and gives insight into the progenitors of supernovae explosions.\nDimension of discretized data: $100$ time-steps of $256\\times 128 \\times 256$ images per trajectory.\nFields available in the data: energy (scalar field), density (scalar field), pressure (Scalar field), velocity (vector field).\nNumber of trajectories: 29 (they are cuts of one long trajectory, long trajectory available on demand).\nEstimated size of the ensemble of all simulations: 570 GB.\nGrid type: spherical coordinates, uniform in $(\\log r, \theta,\\phi)$.  Simulations are done for a portion of a sphere (not the whole sphere), so the simulation volume is like a spherical cake slice.\nInitial and boundary conditions: The temperature at the inner boundary (IB) is first set to equal that of the appropriate radius coordinate in the MESA (1D) model ($400\\~R_\\odot$ and $300\\~R_\\odot$) and the density selected to approximately recover the initial total mass of the star in the simulation ($15.4\\~M_\\odot$ and $14\\~M_\\odot$). \nBetween $300\\~R_\\odot$ and $400\\~R_\\odot$, the initial profile is constructed with the radiative luminosity to be $10^5\\~L_\\odot$, and this is kept fixed in the IB.\nSimulation time-step: 198 days.\nData are stored separated by ($\\Delta t$): units here are sort of arbitrary, $\\Delta t= 8$.\nTotal time range ($t_{min}$ to $t_{max}$):  0, 806 (arbitrary).\nSpatial domain size: $R$ from $300-6700~{\rm R_\\odot}$, θ from $π/4−3π/4$ and $\\phi$ from $0−π$, with $δr/r ≈ 0.01$.\nSet of coefficients or non-dimensional parameters evaluated:\n| Simulation | radius of inner boundary $R_{IB}/R_\\odot$ | radius of outer boundary $R_{OB}/R_\\odot$ | heat source | resolution (r × θ × $\\phi$) | duration | core mass $mc/M\\odot$ | final mass $M_{\rm final}/M_\\odot$ |\n|--|--|--|--|--|--|--|--|\n| Whole simulation (to obtain the 29 trajectories) | 300 | 6700 | fixed L | 256 × 128 × 256 | 5766 days | 10.79 | 12.9 |\nApproximate time to generate the data: 2 months on 80 nodes for each run.\nHardware used to generate the data: 80x NASA Pleiades Skylake CPU nodes.\nWhat phenomena of physical interest are captured in the data: turbulence and convection (inherently 3D processes), variability.\nHow to evaluate a new simulator operating in this space: can it predict behaviour of simulation in convective steady-state, given only perhaps a few snapshots at the beginning of the simulation?\nCaveats: complicated geometry, size of a slice in R varies with R (think of this as a slice of cake, where the parts of the slice closer to the outside have more area/volume than the inner parts), simulation reaches convective steady-state at some point and no longer \"evolves\".", "id": "Red Supergiant Convective Envelope", "primary_category": "physics.flu-dyn", "authors": "['Jared A. Goldberg', 'Yan-Fei Jiang', 'Lars Bildsten']", "updated": null}
+{"dataset": "euler_quadrants", "url": "https://epubs.siam.org/doi/pdf/10.1137/S1064827595291819?casa_token=vkASCwD4WngAAAAA:N0jy0Z6tshitF10_YRTlZzU-P7mAiPFr3v58sw7pmRsZOarAi824-b1CWhOQts1rvaG3YpJisw", "links": "https://epubs.siam.org/doi/pdf/10.1137/S1064827595291819?casa_token=vkASCwD4WngAAAAA:N0jy0Z6tshitF10_YRTlZzU-P7mAiPFr3v58sw7pmRsZOarAi824-b1CWhOQts1rvaG3YpJisw", "abstract": "The evolution can give rise to shocks, rarefaction waves, contact discontinuities, interaction with each other and domain walls.\nDimension of discretized data: 100 timesteps of 512x512 images.\nFields available in the data: density (scalar field), energy (scalar field), pressure (scalar field), momentum (vector field).\nNumber of trajectories: 500 per set of parameters, 10 000 in total.\nEstimated size of the ensemble of all simulations: 5.17 TB.\nGrid type: uniform, cartesian coordinates.\nInitial conditions: Randomly generated initial quadrants.\nBoundary conditions: Periodic or open.\nSimulation time-step: variable.\nData are stored separated by ($\\Delta t$): 0.015s (1.5s for 100 timesteps).\nTotal time range ($t_{min}$ to $t_{max}$): $t_{min} = 0$, $t_{max}=1.5s$.\nSpatial domain size ($L_x$, $L_y$, $L_z$): $L_x = 1$ and  $L_y = 1$.\nS**et of coefficients or non-dimensional parameters evaluated: all combinations of $\\gamma$ constant of the gas: $\\gamma=${1.3,1.4,1.13,1.22,1.33,1.76, 1.365,1.404,1.453,1.597} and boundary conditions: {extrap, periodic}.\nApproximate time to generate the data: 80 hours on 160 CPU cores for all data.\nHardware used to generate the data and precision used for generating the data: Icelake nodes, double precision.\nWhat phenomena of physical interest are catpured in the data: capture the shock formations and interactions. Multiscale shocks.\nHow to evaluate a new simulator operating in this space: the new simulator should predict the shock at the right location and time, and the right shock strength, as compared to a ‘pressure’ gauge monitoring the ‘exact’ solution.", "id": "Euler Multi-quadrants - Riemann problems (compressible, inviscid fluid)", "primary_category": "physics.flu-dyn", "authors": "['Peter D. Lax', 'Xu-Dong Liu']", "updated": null}
+{"dataset": "helmholtz_staircase", "url": null, "links": null, "abstract": "Accurate solution of PDEs near infinite, periodic boundaries poses a numerical challenge due these surfaces serving as waveguides, allowing modes to propagate for long distances from the source. This property makes numerical truncation of the (infinite) solution domain unfeasible, as it would induce large artificial reflections and therefore errors. Periodization (reducing the computational domain to one unit cell) is only possible if the incident wave is also\nperiodic, such as plane waves, but not for nonperiodic sources, e.g. a point source. Computing a high-order accurate scattering solution from a point source, however, would be of scientific interest as it models applications such as remote sensing, diffraction from gratings, antennae, or acoustic/photonic metamaterials. We use a combination of the Floquet—Bloch transform (also known as array scanning method) and boundary integral equation methods to alleviate these challenges and recover the scattered solution as an integral over a family of quasiperiodic solutions parameterized by their on-surface wavenumber. The advantage of this approach is that each of the quasiperiodic solutions may be computed quickly by periodization, and accurately via high-order quadrature.\nDimension of discretized data: $50$ time-steps of \n$1024\\times256$ images.\nFields available in the data:\nreal and imaginary part of accoustic pressure (scalar field), the staircase mask (scalar field, stationary).\nNumber of trajectories: $512$ (combinations of $16$ input parameter $\\omega$ and $32$ source positions $\\mathbf{x}_0$).\nSize of the ensemble of all simulations: 52.4 GB.\nGrid type: uniform.\nInitial conditions: The time-dependence is\nanalytic in this case: $U(t, \\mathbf{x}) = u(\\mathbf{x})e^{-i\\omega t}.$ Therefore any spatial solution may serve as an initial condition.\nBoundary conditions: Neumann conditions (normal\nderivative of the pressure $u$ vanishes, with the normal defined as pointing up from\nthe boundary) are enforced at the boundary.\nSimulation time-step: continuous in time (time-dependence is\nanalytic).\nData are stored separated by ($\\Delta t$): $\\Delta t =\frac{2\\pi}{\\omega N}$, with $N = 50$.\nTotal time range ($t_{min}$ to $t_{max}$): $t_{\\mathrm{min}} = 0$, $t_{\\mathrm{max}} =\n\frac{2\\pi}{\\omega}$.\nSpatial domain size ($L_x$, $L_y$, $L_z$): $-8.0 \\leq x_1 \\leq 8.0$ horizontally, and $-0.5 \\geq x_2 \\geq 3.5$ vertically.\nSet of coefficients or non-dimensional parameters evaluated: $\\omega$={0.06283032, 0.25123038, 0.43929689, 0.62675846, 0.81330465, 0.99856671, 1.18207893, 1.36324313, 1.5412579, 1.71501267, 1.88295798, 2.04282969, 2.19133479, 2.32367294, 2.4331094,  2.5110908}, with the sources coordinates being all combinations of $x$={-0.4, -0.3, -0.2, -0.1, 0, 0.1, 0.2, 0.3, 0.4} and $y$={-0.2, -0.1, 0, 0.1, 0.2, 0.3, 0.4}.\nApproximate time to generate the data: per input parameter: $\\sim 400s$, total: $\\sim 50$ hours.\nHardware used to generate the data: 64 CPU cores.\nWhat phenomena of physical interest are captured in the data: The simulations capture the existence of trapped acoustic waves – modes that are guided along the corrugated surface. They also show that the on-surface wavenumber of trapped modes is different than the frequency of the incident radiation, i.e. they capture the trapped modes’ dispersion relation.\nHow to evaluate a new simulator operating in this space:\nThe (spatial) accuracy of a new simulator/method could be checked by requiring that it conserves flux – whatever the source injects into the system also needs to come out. The trapped modes’ dispersion relation may be another metric, my method generates this to 7-8 digits of accuracy at the moment, but 10-12 digits may also be obtained. The time-dependence learnt by a machine learning algorithm can be compared to the analytic solution $e^{-i\\omega t}$, this can be used to evaluate temporal accuracy.", "id": "Helmholtz equation on a 2D staircase", "primary_category": "physics.flu-dyn", "authors": ["The Well Collaboration"], "updated": null}
+{"dataset": "pattern_formation", "url": null, "links": null, "abstract": "The Gray-Scott equations are a set of coupled reaction-diffusion equations describing two chemical species, $A$ and $B$, whose concentrations vary in space and time. The two parameters $f$ and $k$ control the “feed” and “kill” rates in the reaction. A zoo of qualitatively different static and dynamic patterns in the solutions are possible depending on these two parameters. There is a rich landscape of pattern formation hidden in these equations.\nDimension of discretized data: 1001 time-steps of $128\\times 128$ images.\nFields available in the data: Two chemical species $A$ and $B$.\nNumber of trajectories: 6 sets of parameters, 200 initial conditions per set = 1200.\nEstimated size of the ensemble of all simulations: 153.8 GB.\nGrid type: uniform, cartesian coordinates.\nInitial conditions: Two types of initial conditions generated: random Fourier series and random clusters of Gaussians.\nBoundary conditions: periodic.\nSimulation time-step: 1 second.\nData are stored separated by ($\\Delta t$): 10 seconds.\nTotal time range ($t_{min}$ to $t_{max}$): $t_{min} =0$, $t_{max} = 10,000$.\nSpatial domain size ($L_x$, $L_y$, $L_z$): $x,y\\in[-1,1]$.\nSet of coefficients or non-dimensional parameters evaluated: All simulations used $\\delta_u = 2.10^{-5}$ and $\\delta_v = 1.10^{-5}$.\n\"Gliders\": $f = 0.014, k = 0.054$. \"Bubbles\": $f = 0.098, k =0.057$. \"Maze\": $f= 0.029, k = 0.057$. \"Worms\": $f= 0.058, k = 0.065$. \"Spirals\": $f=0.018, k = 0.051$. \"Spots\": $f= 0.03, k=0.062$.\nApproximate time to generate the data: 5.5 hours per set of parameters, 33 hours total.\nHardware used to generate the data: 40 CPU cores.\nWhat phenomena of physical interest are catpured in the data: Pattern formation: by sweeping the two parameters $f$ and $k$, a multitude of steady and dynamic patterns can form from random initial conditions.\nHow to evaluate a new simulator operating in this space: It would be impressive if a simulator—trained only on some of the patterns produced by a subset of the $(f, k)$ parameter space—could perform well on an unseen set of parameter values $(f, k)$ that produce fundamentally different patterns. Stability for steady-state patterns over long rollout times would also be impressive.", "id": "Pattern formation in the Gray-Scott reaciton-diffusion equations", "primary_category": "physics.flu-dyn", "authors": ["The Well Collaboration"], "updated": null}
+{"dataset": "planetswe", "url": "https://openreview.net/forum?id=RFfUUtKYOG", "links": "https://openreview.net/forum?id=RFfUUtKYOG", "abstract": "The shallow water equations are fundamentally a 2D approximation of a 3D flow in the case where horizontal length scales are significantly longer than vertical length scales. They are derived from depth-integrating the incompressible Navier-Stokes equations. The integrated dimension then only remains in the equation as a variable describing the height of the pressure surface above the flow. These equations have long been used as a simpler approximation of the primitive equations in atmospheric modeling of a single pressure level, most famously in the Williamson test problems. This scenario can be seen as similar to Williamson Problem 7 as we derive initial conditions from the hPa 500 pressure level in ERA5. These are then simulated with realistic topography and two levels of periodicity.\nDimension of discretized data: 3024 timesteps of 256x512 images with \"day\" defined as 24 steps and \"year\" defined as 1008 in model time. \nFields available in the data: height (scalar field), velocity (vector field).\nNumber of trajectories: 40 trajectories of 3 model years.\nEstimated size of the ensemble of all simulations: 185.8 GB.\nGrid type: Equiangular grid, polar coordinates.\nInitial conditions: Sampled from hPa 500 level of [ERA5](https://rmets.onlinelibrary.wiley.com/doi/10.1002/qj.3803), filtered for stable initialization and burned-in for half a simulation year. \nBoundary conditions: Spherical.\nSimulation time-step ($\\Delta t$): CFL-based step size with safety factor of .4. \nData are stored separated by ($\\delta t$): 1 hour in simulation time units.\nTotal time range ($t_{min}$ to $t_{max}$): $t_{min} = 0$, $t_{max} = 3024$.\nSpatial domain size ($L_x$, $L_y$, $L_z$): $\\phi \\in [0, 2 \\pi]$, $\theta \\in [0, \\pi]$. \nSet of coefficients or non-dimensional parameters evaluated: $\nu$ normalized to mode 224. \nApproximate time to generate the data: 45 minutes using 64 icelake cores for one simulation. \nHardware used to generate the data: 64 Icelake CPU cores.\nSpherical geometry and planet-like topography and forcing make for a proxy for real-world atmospheric dynamics where true dynamics are known. The dataset has annual and daily periodicity forcing models to either process a sufficient context length to learn these patterns or to be explicitly time aware. Furthermore, the system becomes stable making this a good system for exploring long run stability of models.", "id": "PlanetSWE", "primary_category": "physics.flu-dyn", "authors": "['Michael McCabe', 'Peter Harrington', 'Shashank Subramanian', 'Jed Brown']", "updated": null}
+{"dataset": "post_neutron_star_merger", "url": null, "links": null, "abstract": "The simulations presented here are axisymmetrized snapshots of full three-dimensional general relativistic neutrino radiation magnetohydrodynamics. The plasma physics is treated with finite volumes with constrained transport for the magnetic field on a curvilinear grid. The system is closed by a tabulated nuclear equation of state assuming nuclear statistical equilibrium (NSE). The radiation field is treated via Monte Carlo transport, which is a particle method. The particles are not included in this dataset, however their effects are visible as source terms on the fluid.\nDimension of discretized data: 181 time-steps of $192 \\times 128 \\times 66$ snapshots.\nFields available in the data: fluid density (scalar field), fluid internal energy (scalar field), electron fraction (scalar field), temperate (scalar field), entropy (scalar field), velocity (vector field), magnetic field (vector field), contravariant tensor metric of space-time (tensor field, no time-dependency).\nA description of fields available in an output file can be found here:\nhttps://github.com/lanl/nubhlight/wiki\nNumber of trajectories: Currently eight full simulations. \nSize of the ensemble of all simulations: 110.1 GB.\nGrid type**: Uniform grid, log-spherical coordinates.\nInitial conditions: Constant entropy torus in hydrostatic equilibrium orbiting a black hole. Black hole mass and spin, as well as torus mass, spin, electron fraction, and entropy vary.\nBoundary conditions: open.\nSimulation time-step: approximately 0.01 in code units. Physical time varies; roughly 147 nanoseconds for fiducial model.\nData are stored separated by ($\\Delta t$): 50 in code units. Physical time varies; roughly 0.6 milliseconds for fiducial model.\nTotal time range ($t_{min}$ to $t_{max}$): 10000 in code units. Physical time varies; roughly 127 milliseocnds for fudicial model\nSpatial domain size ($L_x$, $L_y$, $L_z$): Spherical coordinates. Radius roughly 2 to 1000 in code units. Physical values vary. Outer boundary is at roughly 4000 for fiducial model. Polar angle 0 to pi. Azimuthal angle 0 to 2*pi. Note that the coordinates are curvilinear. In Cartesian space, spacing is logarithmic in radius and there is a focusing of grid lines near the equator.\nSet of coefficients or non-dimensional parameters evaluated: Black hole spin parameter a, ranges 0 to 1. Initial mass and angular momentum of torus. In dimensionless units, evaluated as inner radius Rin and radius of maximum pressure Rmax. Torus initial electron fraction Ye and entropy kb. Black hole mass in solar masses.\nApproximate time to generate the data: Roughly 3 weeks per simulation on 300 cores.\nHardware used to generate the data and precision used for generating the data: Data generated at double precision on several different supercomputers. All calculations were CPU calculations parallelized with a hybrid MPI + OpenMP strategy. 1 MPI rank per socket. Oldest calculations performed on the Los Alamos Badger cluster, now decommissioned. Intel Xeon E5-2695v5 2.1 GHz. 12 cores per socket, 24 core cores per node. Simulations run on 33 nodes. Some newer simulations run on Los Alamos Capulin cluster, now decomissioned. ARM ThunderX2 nodes. 56 cores per node. Simulation run on 33 nodes.\n## Simulation Index\n| Scenario | Shorthand name | Description                                                         |\n|----------|----------------|---------------------------------------------------------------------|\n| 0        | collapsar_hi   | Disk resulting from collapse of massive rapidly rotating star.       |\n| 1        | torus_b10      | Disk inspired by 2017 observation of a neutron star merger. Highest magnetic field strength. |\n| 2        | torus_b30      | Disk inspired by 2017 observation of a neutron star merger. Intermediate magnetic field strength. |\n| 3        | torus_gw170817 | Disk inspired by 2017 observation of a neutron star merger. Weakest magnetic field strength. |\n| 4        | torus_MBH_10   | Disk from black hole-neutron star merger. 10 solar mass black hole.  |\n| 5        | torus_MBH_2p31 | Disk from black hole-neutron star merger. 2.31 solar mass black hole.|\n| 6        | torus_MBH_2p67 | Disk from black hole-neutron star merger. 2.76 solar mass black hole.|\n| 7        | torus_MBH_2p69 | Disk from black hole-neutron star merger. 2.79 solar mass black hole.|\n| 8        | torus_MBH_6    | Disk from black hole-neutron star merger. 6 solar mass black hole.   |\n## General relativistic quantities\nThe core quantity that describes the curvature of spacetime and its\nimpact on a simulation is `['t0_fields']['gcon']` of the HDF5 file. From this other\nquantities can be computed.\n## To reproduce\nThe values in `simulation_parameters.json` are sufficient to reproduce a\nsimulation using [nubhlight](https://github.com/lanl/nubhlight) using\nthe `torus_cbc` problem generator, with one exception. You must\nprovide tabulated equation of state and opacity data. We use the SFHo\nequation of state provided on the\n[stellar collapse website](https://stellarcollapse.org/).\nTabulated neutrino opacities were originally computed for the Fornax\ncode and are not public. However adequate open source substitutes may\nbe generated by the [nulib](http://www.nulib.org/) library.\n## Explanation of simulation parameters\nHere we include, for completeness, a description of the different simulation parameters. which cover the simulation parameters chosen. Their value for each simulation is stored in `simulation_parameters.json`.\n- `B_unit`, the unit of magnetic field strength. Multiplying code quantity by `B_unit` converts the quantity to units of Gauss.\n- `DTd`, dump time cadence.\n- `DTl`, log output time cadence.\n- `DTp`, permanent restart file time cadence.\n- `DTr`, temporary restart file time cadence.\n- `Ledd`, (Photon) Eddington luminosity based on black hole mass.\n- `L_unit`, length unit. Multiplying code quantity by `L_unit` converts it into units of cm.\n- `M_unit`, mass unit. Multiplying code quantity by `M_unit` converts it into units of g.\n- `Mbh`, black hole mass in units of g.\n- `MdotEdd`, (Photon) Eddington accretion rate based on black hole mass.\n- `N1`, number of grid points in X1 (radial) direction.\n- `N2`, number of grid points in X2 (polar) direction.\n- `N3`, number of grid points in X3 (azimuthal) direction.\n- `PATH`, output directory for the original simulation.\n- `RHO_unit`, density unit. Multiplying code quantity by `RHO_unit` converts it into units of g/cm^3.\n- `Reh`, radius of the event horizon in code units.\n- `Rin`, radius of the inner boundary in code units.\n- `Risco`, radius of the innermost stable circular orbit in code units.\n- `Rout_rad`, outer radius of neutrino transport.\n- `Rout_vis`, radius used for 3D volume rendering.\n- `TEMP_unit`, temperature unit. Converts from MeV (code units) to Kelvin.\n- `T_unit`, time unit. Converts from code units to seconds.\n- `U_unit`, energy density unit. Multiplying code quantity by `U_unit` converts it into units of erg/cm^3.\n- `a`, dimensionless black hole spin. \n- `cour`, dimensionless CFL factor used to set the timestep based on the grid spacing.\n- `dx`, array of grid spacing in code coordinates. (Uniform.)\n- `maxnscatt`, maximum number of scattering events per superphoton particle\n- `mbh`, black hole mass in solar masses.\n- `hslope`, `mks_smooth`, `poly_alpha`, `poly_xt` focusing terms used for coordinate transforms\n- `startx`, array of starting coordinate values for `X1`,`X2`,`X3` in code coordinates.\n- `stopx`, array of ending coordinate values for `X1`,`X2`,`X3` in code coordinates.\n- `tf`, final simulation time.\n- `variables` list of names of primitive state vector.\nWhat phenomena of physical interest are catpured in the data: The 2017 detection of the in-spiral and merger of two neutron stars\nwas a landmark discovery in astrophysics. Through a wealth of\nmulti-messenger data, we now know that the merger of these\nultracompact stellar remnants is a central engine of short gamma ray\nbursts and a site of r-process nucleosynthesis, where the heaviest\nelements in our universe are formed. The radioactive decay of unstable\nheavy elements produced in such mergers powers an optical and\ninfra-red transient: The kilonova.\nOne key driver of nucleosynthesis and resultant electromagnetic\nafterglow is wind driven by an accretion disk formed around the\ncompact remnant. Neutrino transport plays a key role in setting the\nelectron fraction in this outflow, thus controlling the\nnucleosynthesis.\nCollapsars are black hole accretion disks formed after the core of a\nmassive, rapidly rotating star collapses to a black hole. These\ndramatic systems rely on much the same physics and modeling as\npost-merger disks, and can also be a key driver of r-processes\nnucleosynthesis.\nHow to evaluate a new simulator operating in this space: The electron fraction of material blown off from the disk is the core\n\"delivarable.\" It determines how heavy elements are synthesized, which\nin turn determines the electromagnetic counterpart as observed on\nEarth. This is the most important piece to get right from an emulator.", "id": "Post neutron star merger", "primary_category": "physics.flu-dyn", "authors": ["The Well Collaboration"], "updated": null}
+{"dataset": "rayleigh_benard", "url": null, "links": null, "abstract": "Rayleigh-Bénard convection involves fluid dynamics and thermodynamics, seen in a horizontal fluid layer heated from below, forming convective cells due to a temperature gradient. With the lower plate heated and the upper cooled, thermal energy creates density variations, initiating fluid motion. This results in Bénard cells, showcasing warm fluid rising and cool fluid descending. The interplay of buoyancy, conduction, and viscosity leads to complex fluid motion, including vortices and boundary layers.\nDimension of discretized data: $200$ timesteps of \n$512\\times128$ images.\nFields are available in the data: buoyancy (scalar vield), pressure (scalar field), velocity (vector field).\nNumber of simulations: $1750$ ($35$ PDE parameters $\\times$ $50$ initial conditions).\nSize of the ensemble of all simulations: 358.4 GB.\nGrid type: uniform, cartesian coordinates.\nInitial conditions: the buoyancy is composed of a dumped noise added to a linear background  $b(t=0) = (Lz-z)\\times\\delta b_0 + z(Lz-z) \\times\\epsilon$ where $\\epsilon$ is a Gaussian white noise of scale $10^{-3}$.\nThe other fields $u$ and $p$ are initialized to $0$.\nBoundary conditions: periodic on the horizontal direction, Dirichlet conditions on the vertical direction.\nSimulation time-step: 0.25.\nTotal time range ($t_{min}$ to $t_{max}$): $t_{min} = 0$, $t_{max} = 50$.\nSpatial domain size: $0 \\leq x \\leq 4$ horizontally, and $0 \\leq z \\leq 1$ vertically.\nSet of coefficients or non-dimensional parameters evaluated: $\text{Rayleigh}\\in[1e6,1e7,1e8,1e9,1e10], \text{Prandtl}\\in[0.1,0.2,0.5,1.0,2.0,5.0,10.0]$. For initial conditions $\\delta b_0\\in[0.2,0.4,0.6,0.8,1.0]$, the seed used to generate the initial Gaussian white noise are $40,\\ldots,49$.\nApproximate time to generate the data: per input parameter** from $\\sim6\\,000s$ to $\\sim 50\\,000s$ (high Rayleigh numbers take longer), total: $\\sim 60$ hours.\nHardware used to generate the data and precision used for generating the data: 12 nodes of 64 CPU cores with 8 processes per node, in single precision.\nRayleigh-Bénard convection datasets offer valuable insights into fluid dynamics under thermal gradients, revealing phenomena like thermal plumes and turbulent eddies. Understanding these dynamics is crucial for applications in engineering and environmental science.", "id": "Rayleigh Bénard convection", "primary_category": "physics.flu-dyn", "authors": ["The Well Collaboration"], "updated": null}
+{"dataset": "rayleigh_taylor_instability", "url": "https://www.researchgate.net/publication/243660629_Reynolds_number_effects_on_Rayleigh-Taylor_instability_with_possible_implications_for_type_Ia_supernovae", "links": "https://www.researchgate.net/publication/243660629_Reynolds_number_effects_on_Rayleigh-Taylor_instability_with_possible_implications_for_type_Ia_supernovae", "abstract": "We consider the Rayleigh-Taylor instability for a range of Atwood numbers and initial perturbations, all of which have a log—normal horizontal energy spectrum with random phase. The dataset examines how varying the mean, standard deviation and the disparity of the random phase effects the transition to and statistics of the ensuing turbulent flow.\nDimension of discretized data: 60 time-steps of $128\\times 128\\times 128$ cubes.\nFields available in the data: Density (scalar field), velocity (vector field).\nNumber of trajectories: 45 trajectories.\nEstimated size of the ensemble of all simulations: 255.6 GB.\nGrid type: uniform grid, cartesian coordinates.\nInitial conditions: Initial conditions have been set by imposing a log—normal profile for the shape of energy spectrum in wavenumber space, such that:\n$$A(k) = \frac{1}{k\\sigma\\sqrt{2\\pi}} \\exp\\Big(-\frac{(\\ln (k) - \\mu)^2}{2\\sigma^2}\\Big) \\quad\textrm{with}\\quad k = \\sqrt{k^2_x+k^2_y}$$\nwhere $\\mu$ is the mean and $\\sigma$ is the standard deviation of the profile. Furthermore, we have imposed a random phase to the corresponding complex Fourier component (i.e. a random value for the argument of the complex Fourier component) between zero and a varied maximum ($\\phi_{max}$), finally after Fourier transforming to physical space the mean of the resulting profile is normalized to $3.10^5$ to ensure comparable power. \nBoundary conditions:** Periodic boundary conditions on sides walls and slip conditions on the top and bottom walls.\nSimulation time-step: $\\Delta t$ is set such that the maximum Courant number is $\frac12(CFL_{max}=0.5)$. Therefore, the time step decreases as the flow accelerates.\nData are stored separated by ($\\Delta t$):  different according to At number.\nThe time difference between frames varies as the flow accelerates, thus the largest occur at the beginning of the simulation ($\\delta t \\sim 5s$) and the smallest at the end ($\\delta t \\sim 0.1s$).\nTotal time range ($t_{min}$ to $t_{max}$): Varies from $t_{min}=0$ to $t_{max}$ between $\\sim 30s$ and $\\sim 100s$, depending on Atwood number.\nSpatial domain size ($L_x$, $L_y$, $L_z$): $[0,1]\\times[0,1]\\times[0,1]$.\nSet of coefficients or non-dimensional parameters evaluated: We run simulations with 13 different initializations for five different Atwood number $At\\in {\frac34, \frac12, \frac14, \frac18, \frac{1}{16}}$. The first set on initial conditions considers varying the mean $\\mu$ and standard deviation $\\sigma$ of the profile $A(k)$ with $\\mu\\in{1, 4, 16}$ and $\\sigma\\in{\frac14, \frac12, 1}$, the phase (argument of the complex Fourier component) $\\phi$ was set randomly in the range $[0,2\\pi)$. The second set of initial conditions considers a fixed mean ($\\mu=16$) and standard deviation ($\\sigma =0.25$) and a varieed range of random phases (complex arguments $\\phi\\in[0,\\phi_{max}$)) given to each Fourier component. The four cases considered are specified by $\\phi_{max}\\in { \frac{\\pi}{128}, \frac{\\pi}{8}, \frac{\\pi}{2}, \\pi}$. \nApproximate time to generate the data: 1 hour on 128 CPU cores for 1 simulation. 65 hours on 128 CPU cores for all simulations.\nHardware used to generate the data: 128 CPU core on the Ocre supercomputer at CEA, Bruyères-le-Châtel, France.\nWhat phenomena of physical interest are catpured in the data: In this data there are three key aspects of physical interest. Firstly, impact of coherence on otherwise random initial conditions. Secondly, the effect of the shape of the initial energy spectrum on the structure of the flow. Finally, the transition from the Boussinesq to the non-Boussinesq regime where the mixing width transitions from symmetric to asymmetric growth.  \nHow to evaluate a new simulator operating in this space:\nFrom a fundamental standpoint we, would expect the density field to be advected and mixed rather than created or destroyed to give appropriate statistics. From a qualitative perspective, given that the underlying simulations are of comparable spatial resolution to the simulations run by the alpha group (Dimonte et. al. 2003) we would consider a good emulator to produce a comparable value for α as reported in their paper for an appropriately similar set of initial conditions. This parameter is derived by considering the flow after the initial transient. At this stage, the width of the turbulent mixing zone, $L$, is self-similar and grows as $L= \u0007lpha \\* At \\* g \\* t^2$. They reported a value of $\u0007lpha$=0.025±0.003. In addition, during this self-regime, we would expect to observe energy spectra with a similar shape to those reported in Cabot and Cook 2006, specifically exhibiting an appropriate $k^{-\frac53}$ cascade. From a structural perspective, we would expect that for an initialization with a large variety of modes in the initial spectrum to observe a range of bubbles and spikes (upward and downward moving structures), whereas in the other limit (where this only on mode in the initial spectrum) we expect to observe a single bubble and spike.  In addition, a good emulator would exhibit symmetric mixing with for low Atwood numbers in the Boussinesq regime (defined as $At$ < 0.1 by Andrews and Dalziel 2010) and asymmetries in the mixing with for large Atwood number.", "id": "Rayleigh-Taylor instability", "primary_category": "physics.flu-dyn", "authors": ["William  H. Cabot", "Andrew W. Cook"], "updated": null}
+{"dataset": "shear_flow", "url": null, "links": null, "abstract": "Shear flow are a type of fluid characterized by the continuous deformation of adjacent fluid layers sliding past each other with different velocities. This phenomenon is commonly observed in various natural and engineered systems, such as rivers, atmospheric boundary layers, and industrial processes involving fluid transport.\nThe dataset explores a 2D periodic shearflow governed by incompressible Navier-Stokes equation.\nDimension of discretized data: $200$ time-steps of $128\\times256$ images.\nFields available in the data: tracer (scalar field), velocity (vector field), pressure (scalar field).\nNumber of simulations: $1120$ ($28$ PDE parameters $\\times$ $40$ initial conditions).\nSize of the ensemble of all simulations: 114.7 GB.\nGrid type: uniform, cartesian coordinates.\nInitial conditions: the shear field $u_1$ is composed of $n_\text{shear}$ shears uniformly spaced along the $z$ direction. Each shear is implemented with a tanh (hyperbolic tangent) $\text{tanh}(5\frac{y-y_k}{n_\text{shear}w})$ where $z_k$ is the vertical position of the shear and $w$ is a width factor.\nThe velocity field $u_2$ is composed of sinusoids along the $x$ direction located at the shear. These sinusoids have an exponential decay away from the shear in the $y$ direction $\text{sin}(n_\text{blobs}\\pi x)\\,e^{\frac{25}{w^2}|y-y_k|^2}$.\nThe tracer matches the shear at initialization. The pressure is initialized to zero.\nThe initial condition is thus indexed by $n_\text{shear},n_\text{blobs},w$.\nBoundary conditions: periodic.\nSimulation time-step: 0.1.\nTotal time range ($t_{min}$ to $t_{max}$): $t_{min} = 0$, $t_{max} = 20$.\nSpatial domain size: $0\\leq x \\leq 1$ horizontally, and $-1 \\leq y \\leq 1$ vertically.\nSet of coefficients or non-dimensional parameters evaluated: $\text{Reynolds}\\in[1e4, 5e4, 1e5, 5e5], \text{Schmidt}\\in[0.1, 0.2, 0.5, 1.0, 2.0, 5.0, 10.0]$. For initial conditions $n_\text{shear}\\in[2,4]$, $n_\text{blobs}\\in[2,3,4,5]$, $w\\in[0.25, 0.5, 1.0, 2.0, 4.0]$.\nApproximate time to generate the data: per input parameter: $\\sim 1500s$, total: $\\sim 5$ hours.\nHardware used to generate the data and precision used for generating the data: 7 nodes of 64 CPU cores each with 32 tasks running in parallel on each node, in single precision.\nShear flow are non-linear phenomena arrising in fluid mechanics and turbulence.\nPredicting the behavior of the shear flow under different Reynolds and Schmidt numbers is essential for a number of applications in aerodynamics, automotive, biomedical. \nFurthermore, such flow are unstable at large Reynolds number.", "id": "Periodic shear flow", "primary_category": "physics.flu-dyn", "authors": ["The Well Collaboration"], "updated": null}
+{"dataset": "supernova_explosion_128", "url": null, "links": null, "abstract": "The simulations solve an explosion inside a compression of a monatomic ideal gas, which follows the equation of state with the specific heat ratio $\\gamma=5/3$.\nThe gas in these simulations mocks interstellar medium in the Milky Way Galaxy.\nAt the beginning of the simulations, the thermal energy of a supernova is dumped at the center of the simulation box.\nThe hot ($\\sim 10^7$ K) gas is immediately accelerated and makes the blastwave.\nBecause velocities of the hot gas become supersonic, much fine resolution and small timestep are required to resolve the dynamics.\nThe physical quantities are also distributed in seven orders of magnitude, which requires a large number of simulation steps.\nDimension of discretized data** $59$ time-steps of  $128\\times 128\\times 128$ cubes.\nFields available in the data:\nPressure (scalar field), density (scalar field), temperature(scalar field), velocity (tensor field).\nNumber of trajectories: 260.\nEstimated size of the ensemble of all simulations: 754 GB\nGrid type: uniform, cartesian coordinates.\nInitial conditions: $820$ random seeds generated using https://github.com/amusecode/amuse/blob/main/src/amuse/ext/molecular_cloud.py (Virialized isothermal gas sphere with turbulence following the velocity spectrum $E(k) \\propto k^{-2}$, which is Burgers turbulence ([Burgers 1948](https://www.sciencedirect.com/science/article/abs/pii/S0065215608701005) and [Kupilas+2021](https://doi.org/10.1093/mnras/staa3889) for reference ))\nBoundary conditions: open.\nData are stored separated by ($\\Delta t$): $100$ ~ $10 000$ years (variable timesteps).\nTotal time range ($t_{min}$ to $t_{max}$): $0$ yr to $0.2$ Myr.\nSpatial domain size ($L_x$, $L_y$, $L_z$): 60 pc.\nSet of coefficients or non-dimensional parameters evaluated: Initial temperature $T_0$=\\{100K\\}, Initial number density of hydrogen $\rho_0=$\\{44.5/cc\\}, metallicity (effectively strength of cooling) $Z=\\{Z_0\\}$.\nApproximate time to generate the data (CPU hours):\n|  0.1 M $\\odot$ |\n|:----------:|\n|  $3500$ |\nHardware used to generate the data and precision used for generating the data: up to 1040 CPU cores per run.\nWhat phenomena of physical interest are catpured in the data:\nThe simulations are designed as an supernova explosion, which is the explosion at the last moment of massive stars, in a high-density starforming molecular cloud with a large density contrast. An adiabatic compression of a monatomic ideal gas is assumed.\nTo mimic the explosion, the huge thermal energy ($10^{51}$ erg) is injected at the center of the calculation box and going to make the blastwave, which sweeps out the ambient gas and shells called as supernova feedback. These interactions between supernovae and surrounding gas are interesting because stars are formed in dense and cold regions.\nHowever, calclatig the propagation of blastwaves requires tiny timesteps to calculate and numerous integration steps. When supernova feedback is incorporated in a galaxy simulation, some functions fitted using local high resolution simulations have been used.\nHow to evaluate a new simulator operating in this space:\nIn context of galaxy simulations, the time evolution of thermal energy and momentum are important. We note that those physical quantities are not necessarily conserved because radiative cooling and heating are considered and thermal energy is seamlessly being converted into momentum.", "id": "Supernova Explosion in Turbulent Interstellar medium in galaxies", "primary_category": "physics.flu-dyn", "authors": ["The Well Collaboration"], "updated": null}
+{"dataset": "supernova_explosion_64", "url": null, "links": null, "abstract": "The simulations solve an explosion inside a compression of a monatomic ideal gas, which follows the equation of state with the specific heat ratio $\\gamma=5/3$.\nThe gas in these simulations mocks interstellar medium in the Milky Way Galaxy.\nAt the beginning of the simulations, the thermal energy of a supernova is dumped at the center of the simulation box.\nThe hot ($\\sim 10^7$ K) gas is immediately accelerated and makes the blastwave.\nBecause velocities of the hot gas become supersonic, much fine resolution and small timestep are required to resolve the dynamics.\nThe physical quantities are also distributed in seven orders of magnitude, which requires a large number of simulation steps.\nDimension of discretized data** $59$ time-steps of  $64\\times 64\\times 64$ cubes.\nFields available in the data:\nPressure (scalar field), density (scalar field), temperature(scalar field), velocity (tensor field).\nNumber of trajectories: 740.\nEstimated size of the ensemble of all simulations: 268.2 GB.\nGrid type: uniform, cartesian coordinates.\nInitial conditions: $820$ random seeds generated using https://github.com/amusecode/amuse/blob/main/src/amuse/ext/molecular_cloud.py (Virialized isothermal gas sphere with turbulence following the velocity spectrum $E(k) \\propto k^{-2}$, which is Burgers turbulence ([Burgers 1948](https://www.sciencedirect.com/science/article/abs/pii/S0065215608701005) and [Kupilas+2021](https://doi.org/10.1093/mnras/staa3889) for reference).\nBoundary conditions: open.\nData are stored separated by ($\\Delta t$): $100$ ~ $10,000$ years (variable timesteps). [CHECK]\nTotal time range ($t_{min}$ to $t_{max}$): $0$ yr to $0.2$ Myr.\nSpatial domain size ($L_x$, $L_y$, $L_z$): 60 pc.\nSet of coefficients or non-dimensional parameters evaluated: Initial temperature $T_0$=\\{100K\\}, Initial number density of hydrogen $\rho_0=$\\{44.5/cc\\}, metallicity (effectively strength of cooling) $Z=\\{Z_0\\}$.\nApproximate time to generate the data (CPU hours):\n| 1M $_\\odot$ | 0.1 M $\\odot$ |\n|:----------:|:----------:|\n| $300$ | $3500$ |\nHardware used to generate the data and precision used for generating the data: up to 1040 CPU cores per run.\nWhat phenomena of physical interest are catpured in the data:\nThe simulations are designed as an supernova explosion, which is the explosion at the last moment of massive stars, in a high-density starforming molecular cloud with a large density contrast. An adiabatic compression of a monatomic ideal gas is assumed.\nTo mimic the explosion, the huge thermal energy ($10^{51}$ erg) is injected at the center of the calculation box and going to make the blastwave, which sweeps out the ambient gas and shells called as supernova feedback. These interactions between supernovae and surrounding gas are interesting because stars are formed in dense and cold regions.\nHowever, calclating the propagation of blastwaves requires tiny timesteps to calculate and numerous integration steps. When supernova feedback is incorporated in a galaxy simulation, some functions fitted using local high resolution simulations have been used.\nHow to evaluate a new simulator operating in this space:\nIn context of galaxy simulations, the time evolution of thermal energy and momentum are important. We note that those physical quantities are not necessarily conserved because radiative cooling and heating are considered and thermal energy is seamlessly being converted into momentum.", "id": "Supernova Explosion in Turbulent Interstellar medium in galaxies", "primary_category": "physics.flu-dyn", "authors": ["The Well Collaboration"], "updated": null}
+{"dataset": "turbulence_gravity_cooling", "url": "https://academic.oup.com/mnras/article/526/3/4054/7316686", "links": "https://academic.oup.com/mnras/article/526/3/4054/7316686", "abstract": "These simulations are a turbulent fluid with gravity modeling interstellar medium in galaxies. These fluids make dense filaments, which will form new stars. The timescale and frequency of making new filaments are varied depending on the strength of cooling. It is parametrized by the amount of metal (metallicity), density, and temperature.\nDimension of discretized data: $50$ time-steps of  $64\\times 64\\times 64$ cubes.\nFields available in the data: Pressure (scalar field), density (scalar field), temperature(scalar field), velocity (tensor field).\nNumber of trajectories: 2700 (27 parameters sets $\\times$ 100 runs).\nEstimated size of the ensemble of all simulations: 829.4 GB.\nGrid type: uniform, cartesian coordinates.\nInitial conditions: $2700$ random seeds generated using https://github.com/amusecode/amuse/blob/main/src/amuse/ext/molecular_cloud.py (Virialized isothermal gas sphere with turbulence following the velocity spectrum $E(k) \\propto k^{-2}$, which is Burgers turbulence (Burgers 1948 and Kupilas+2021 for reference )).\nBoundary conditions: open.\nSimulation time-step: $2,000$ ~ $10,000$ years (variable timesteps).\nData are stored separated by ($\\Delta t$): 0.02 free fall time.\nTotal time range ($t_{min}$ to $t_{max}$): 1 Free Fall time (= $L^3/GM$ ); $L=(\rho / \rho_0)^{1/3} \\times 60$ pc, $\rho_0=44.5/\rm{cc}$, $M=1,000,000$ M $_\\odot$.\nSpatial domain size ($L_x$, $L_y$, $L_z$):\n|           | Domain Length ($L$) | Free Fall Time | Snapshot ($\\delta t$) |\n|----------|:----------:|:----------:|:----------:|\n| **Dense (44.5 cm $^{-3}$)** | 60 pc | 6.93 Myr | 0.14 Myr |\n| **Moderate (4.45 cm $^{-3}$)** | 129 pc | 21.9 Myr |0.44 Myr |\n| **Sparse (0.445 cm $^{-3}$)** | 278 pc | 69.3 Myr | 1.4 Myr |\nSet of coefficients or non-dimensional parameters evaluated: Initial temperature $T_0$=\\{10K, 100K, 1000K\\}, Initial number density of hydrogen $\rho_0=$\\{44.5/cc, 4.45/cc, 0.445/cc\\}, metallicity (effectively strength of cooling) $Z=\\{Z_0, 0.1Z_0, 0\\}$.\nApproximate time to generate the data: $600,000$ node hours for all simulations.\n#### For dense dataset (CPU hours)\n|           | Strong (1Z $_\\odot$) | Weak (0.1 Z $_\\odot$) | Adiabatic (0 Z $_\\odot$) |\n|----------:|----------:|----------:|----------:|\n| **$10$ K** | $240$  | $167$ | $77$ |\n| **$100$ K** | $453$ | $204$  | $84$ |\n| **$1000$ K** | $933$ | $186$  | $46$ |\n#### For moderate dataset (CPU hours)\n|           | Strong (1Z $_\\odot$) | Weak (0.1 Z $_\\odot$) | Adiabatic (0 Z $_\\odot$) |\n|----------:|----------:|----------:|----------:|\n| **$10$ K** | $214$  | $75$ | $62$ |\n| **$100$ K** | $556$ | $138$  | $116$ |\n| **$1000$ K** | $442$ | $208$  | $82$ |\n#### For sparse dataset (CPU hours)\n|           | Strong (1Z $_\\odot$) | Weak (0.1 Z $_\\odot$) | Adiabatic (0 Z $_\\odot$) |\n|----------:|----------:|----------:|----------:|\n| **$10$ K** | $187$  | $102$ | $110$ |\n| **$100$ K** | $620$ | $101$  | $92$ |\n| **$1000$ K** | $286$ | $129$  | $93$ |\nHardware used to generate the data and precision used for generating the data: up to 1040 CPU cores per run.\nWhat phenomena of physical interest are catpured in the data:\nGravity, hydrodynamics and radiative cooling/heating are considered in the simulations. Radiative cooling/heating is parameterized with metallicity, which the ratio of heavier elements than helium. The larger and metallicity corresponds to the later and early stage of galaxies and universe, respectively.\nIt also affects the time scale of cooling/heating and star formation rate. For instance, star formation happens at dense and cold region. With the strong cooling/heating rate, dense regions are quickly cooled down and generates new stars. Inversely, in the case of a weak cooling/heating, when gas is compressed, it is heated up and prevent new stars from being generated.\nIn the case of cold gas with strong cooling/heating, it can easily make dense regions, which require small timesteps and a lot of integration steps. That makes it difficult to get the resolution higher.\nHow to evaluate a new simulator operating in this space:\nThe new simulator should be able to detect potential regions of star formation / potential number of newborn stars, because star formation regions are very dense and need very small timesteps, which results in a massive number of calculation steps.", "id": "Turbulent Interstellar medium in galaxies", "primary_category": "physics.flu-dyn", "authors": "['Keiya Hirashima', 'Kana Moriwaki', 'Michiko S Fujii', 'Yutaka Hirai', 'Takayuki R Saitoh', 'Junichiro Makino']", "updated": null}
+{"dataset": "turbulent_radiative_layer_2D", "url": "https://iopscience.iop.org/article/10.3847/2041-8213/ab8d2c/pdf", "links": "https://iopscience.iop.org/article/10.3847/2041-8213/ab8d2c/pdf", "abstract": "In this simulation, there is cold, dense gas on the bottom and hot dilute gas on the top. They are moving relative to each other at highly subsonic velocities. This set up is unstable to the Kelvin Helmholtz instability, which is seeded with small scale noise that is varied between the simulations. The hot gas and cold gas are both in thermal equilibrium in the sense that the heating and cooling are exactly balanced. However, once mixing occurs as a result of the turbulence induced by the Kelvin Helmholtz instability the intermediate temperatures become populated. This intermediate temperature gas is not in thermal equilibrium and cooling beats heating. This leads to a net mass flux from the hot phase to the cold phase. This process occurs in the interstellar medium, and in the Circum-Galactic medium when cold clouds move through the ambient, hot medium. By understanding how the total cooling and mass transfer scale with the cooling rate we are able to constrain how this process controls the overall phase structure, energetics and dynamics of the gas in and around galaxies.\nDimension of discretized data: 101 timesteps of 384x128 images.\nFields available in the data: Density (scalar field), pressure (scalar field), velocity (vector field).\nNumber of trajectories: 90 (10 different seeds for each of the 9 $t_{cool}$ values).\nEstimated size of the ensemble of all simulations: 6.9 GB.\nGrid type: uniform, cartesian coordinates.\nInitial conditions: Analytic, described in the [paper](https://ui.adsabs.harvard.edu/abs/2020ApJ...894L..24F/abstract).\nBoundary conditions: Periodic in the x-direction, zero-gradient for the y-direction.\nSimulation time-step ($\\Delta t$): varies with $t_{cool}$. Smallest $t_{cool}$ has $\\Delta t = 1.36\\times10^{-2}$ and largest $t_{cool}$ has $\\Delta t = 1.74\\times10^{-2}$. Not that this is not in seconds. This is in dimensionless simulation time.\nData are stored separated by ($\\delta t$): 1.597033 in simulation time.\nTotal time range ($t_{min}$ to $t_{max}$): $t_{min} = 0$, $t_{max} = 159.7033$.\nSpatial domain size ($L_x$, $L_y$, $L_z$): $x \\in [-0.5, 0.5]$, $y \\in [-1, 2]$ giving $L_x = 1$ and $L_y = 3$.\nSet of coefficients or non-dimensional parameters evaluated: $t_{cool} = \\{0.03, 0.06, 0.1, 0.18, 0.32, 0.56, 1.00, 1.78, 3.16\\}$. \nApproximate time to generate the data: 84 seconds using 48 cores for one simulation. 100 CPU hours for everything.\nHardware used to generate the data: 48 CPU cores.\nWhat phenomena of physical interest are catpured in the data:\n-\tThe mass flux from hot to cold phase.\n-\tThe turbulent velocities.\n-\tAmount of mass per temperature bin (T = press/dens).\nHow to evaluate a new simulator operating in this space: See whether it captures the right mass flux, the right turbulent velocities, and the right amount of mass per temperature bin.", "id": "Turbulent Radiative Layer - 2D", "primary_category": "physics.flu-dyn", "authors": "['Drummond B. Fielding', 'Eve C. Ostriker', 'Greg L. Bryan', 'Adam S. Jermyn']", "updated": null}
+{"dataset": "turbulent_radiative_layer_3D", "url": "https://iopscience.iop.org/article/10.3847/2041-8213/ab8d2c/pdf", "links": "https://iopscience.iop.org/article/10.3847/2041-8213/ab8d2c/pdf", "abstract": "In this simulation, there is cold, dense gas on the bottom and hot dilute gas on the top. They are moving relative to each other at highly subsonic velocities. This set up is unstable to the Kelvin Helmholtz instability, which is seeded with small scale noise that is varied between the simulations. The hot gas and cold gas are both in thermal equilibrium in the sense that the heating and cooling are exactly balanced. However, once mixing occurs as a result of the turbulence induced by the Kelvin Helmholtz instability, the intermediate temperatures become populated. This intermediate temperature gas is not in thermal equilibrium, and cooling beats heating. This leads to a net mass flux from the hot phase to the cold phase. This process occurs in the interstellar medium, and in the Circum-Galactic medium when cold clouds move through the ambient, hot medium. By understanding how the total cooling and mass transfer scale with the cooling rate, we are able to constrain how this process controls the overall phase structure, energetics and dynamics of the gas in and around galaxies.\nDimension of discretized data: 101 timesteps of 256x128x128 arrays.\nFields available in the data: Density (scalar field), pressure (scalar field), velocity (vector field).\nNumber of trajectories: 90 trajectories (10 different seeds for each of the 9 $t_{cool}$ variations).\nEstimated size of the ensemble of all simulations: 744.6 GB.\nGrid type: uniform, cartesian coordinates.\nInitial conditions: Analytic, described in the [paper](https://ui.adsabs.harvard.edu/abs/2020ApJ...894L..24F/abstract).\nBoundary conditions: periodic for the 128x128 directions ($x,y$), and zero-gradient for the 256 direction ($z$).\nSimulation time-step: varies with $t_{cool}$. Smallest $t_{cool}$ is $1.32.10^{-2}$, largest $t_{cool}$ is $1.74.10^{-2}$. This is not in seconds, as this is a dimensionless simulation time. To convert, the code time is $L_{box}/cs_{hot}$, where $L_{box}$= 1 parsec and cs_{hot}=100km/s.\nData are stored separated by ($\\Delta t$): data is separated by intervals of simulation time of 2.661722.\nTotal time range ($t_{min}$ to $t_{max}$): $t_{min} = 0$, $t_{max} = 266.172178$.\nSpatial domain size ($L_x$, $L_y$, $L_z$): $x,y\\in\\[-0.5,0.5\\]$, $z\\in\\[-1,1\\]$.\nSet of coefficients or non-dimensional parameters evaluated: $t_{cool} = \\{0.03, 0.06, 0.1, 0.18, 0.32, 0.56, 1.00, 1.78, 3.16\\}$.\nApproximate time to generate the data: $34,560$ CPU hours for all simulations.\nHardware used to generate the data: each simulation was generated on a 128 core \"Rome\" node.\nWhat phenomena of physical interest are catpured in the data: Capte the mass flux from hot to cold phase. Capture turbulent velocities. Capture the amount of mass per temperature bin ($T = \frac{P}{\rho}$).\nHow to evaluate a new simulator operating in this space: Check whether the above physical phenomena are captured by the algorithm.", "id": "Turbulent Radiative Mixing Layers - 3D", "primary_category": "physics.flu-dyn", "authors": "['Drummond B. Fielding', 'Eve C. Ostriker' , 'Greg L. Bryan', 'Adam S. Jermyn']", "updated": null}
+{"dataset": "viscoelastic_instability", "url": "https://www.cambridge.org/core/services/aop-cambridge-core/content/view/D63B7EDB638451A6FC2FBBFDA85E1BBD/S0022112024000508a.pdf/multistability-of-elasto-inertial-two-dimensional-channel-flow.pdf", "links": "https://www.cambridge.org/core/services/aop-cambridge-core/content/view/D63B7EDB638451A6FC2FBBFDA85E1BBD/S0022112024000508a.pdf/multistability-of-elasto-inertial-two-dimensional-channel-flow.pdf", "abstract": "Elasto-inertial turbulence (EIT) is a recently discovered two-dimensional chaotic flow state observed in dilute polymer solutions. Two-dimensional direct numerical simulations show (up to) four coexistent attractors: the laminar state (LAM), a steady arrowhead regime (SAR), Elasto-inertial turbulence (EIT) and a ‘chaotic arrowhead regime’ (CAR). The SAR is stable for all parameters considered here, while the final pair of (chaotic) flow states are visually very similar and can be distinguished only by the presence of a weak polymer arrowhead structure in the CAR regime. Both chaotic regimes are maintained by an identical near-wall mechanism and the weak arrowhead does not play a role. The data set includes snapshots on the four attractors as well as two edge states. An edge state is an unstable state that exists on the boundary between two basins of attractors, the so-called edge manifold. Edge states have a single unstable direction out of the manifold and are relevant since the lie exactly on the boundary separating qualitatively different behaviours of the flow. The edge states in the present data set are obtained through edge tracking between the laminar state and EIT and between EIT and SAR.\nDimension of discretized data: \n- EIT: 34 trajectories with 60 timesteps, 512x512 images (chaotic solution). \n- CAR: 39 trajectories with 60 timesteps, 512x512 images (chaotic solution).\n- SAR: 20 trajectories with 20 timesteps, 512x512 images (simple periodic solutions). \n- Transition to chaos between EIT and SAR: 36 snapshots with 20 timesteps of 512x512 images. \n- Transition to non-chaotic state between EIT and SAR: 38 snapshots with 20 timesteps of 512x512 images. \n- Transition to chaos between EIT and Laminar: 43 snapshots with 20 timesteps of 512x512 images. \n- Transition to non-chaotic state between EIT and Laminar: 49 snapshots with 20 timesteps of 512x512 images.\nFields available in the data: pressure (scalar field), velocity (vector field), positive conformation tensor ( $c\\_{xx}^{\\*}, c^{\\*}\\_{yy},, c^{\\*}\\_{xy}$ are in tensor fields, $c^{\\*}\\_{zz}$ in scalar fields).\nNumber of trajectories: 260 trajectories.\nEstimated size of the ensemble of all simulations: 66 GB.\nGrid type: uniform cartesian coordinates.\nInitial conditions:\n- Edge trajectory: linear interpolation between a chaotic and a non-chaotic state. \n- SAR: continuation of the solution obtained through a linear instability at a different parameter set using time-stepping. \n- EIT: laminar state + blowing and suction at the walls. \n- CAR: SAR + blowing and suction at the walls.\nBoundary conditions: no slip conditions for the velocity ( $(u^\\*,v^\\*)=(0,0)$ ) at the wall and $\\epsilon=0$ at the wall for the equation for $\\mathbf{C^*}$.\nSimulation time-step: various in the different states, but typically $\\sim 10^{-4}$.\nData are stored separated by ($\\Delta t$): various at different states, but typically 1.\nTotal time range ($t_{min}$ to $t_{max}$): depends on the simulation.\nSpatial domain size ($L_x$, $L_y$, $L_z$): $0 \\leq x \\leq 2\\pi$, $-1 \\leq y \\leq 1$.\nSet of coefficients or non-dimensional parameters evaluated: Reynold number $Re=1000$, Weissenberg number $Wi = 50$, $\beta =0.9$, $\\epsilon=2.10^{-6}$, $L_{max}=70$.\nApproximate time to generate the data: 3 months to generate all the data. It takes typically 1 day to generate $\\sim 50$ snapshots.\nHardware used to generate the data: typically 32 or 64 cores.\nWhat phenomena of physical interest are catpured in the data: The phenomena of interest in the data is: (i) chaotic dynamics in viscoelastic flows in EIT and CAR. Also note that they are separate states. (ii) multistability for the same set of parameters, the flow has four different behaviours depending on the initial conditions.\nHow to evaluate a new simulator operating in this space:\nA new simulator would need to capture EIT/CAR adequately for a physically relevant parameter range.", "id": "Multistability of viscoelastic fluids in a 2D channel flow", "primary_category": "physics.flu-dyn", "authors": "['Miguel Beneitez', 'Jacob Page', 'Yves Dubief', 'Rich R. Kerswel']", "updated": null}

data/results/.gitignore

@@ -0,0 +1,3 @@
+*
+!README.md
+!.gitignore

data/results/README.md

@@ -0,0 +1,19 @@
+This folder contains all the resulting model generations for tagged concepts once the raw data is passed to the model. 
+
+It stores data folders named after:
+- the mode of the run: `eval` when running on dev set OR `new` when running the model on the test set 
+- the order of selecting few-shot examples from the train set: "random", "first", "last", "middle", "distributed", "longest", "shortest"
+- the format of outputs produced: `json` OR `readable`
+- a UUID from randomized alphanumeric symbols
+
+
+Refer to [manual/README.md](../manual/README.md) to undestand the `train`, `dev` and `test` sets.
+
+
+Files in each folder in `results`:
+
+1. `metrics.json`: contains the metrics of the run. For `new` mode, only the average and total time is reported as there is ground truth to measure comparison metrics. 
+2. `predicted_responses.txt`: contains the raw model responses in text format which is then processed
+3. `predictions.json`: contains a JSON object of all the model predictions at the sentence level for each sentence in the test or dev set, extracted from the model response text
+4. `prompts.txt`: contains all the prompts (few-shot examples prefix + input sentences from paper titles and abstracts), model responses, the extracted predictions from model responses and the true tags
+5. `logs/log.txt`: the logger outputs

data/results/best_config.json

@@ -0,0 +1,12 @@
+{
+    "metrics": {
+        "avg_pred_response_time_per_sentence": 5.3649,
+        "total_time": 718.9355,
+        "precision": 0.4161,
+        "recall": 0.3926,
+        "f1": 0.404,
+        "union_precision": 0.5654,
+        "union_recall": 0.5245,
+        "union_f1": 0.5442
+    },
+    "config": 

data/results/eval_random_readable_zoOLWDdW_20240716_040746/ground_truth.json

@@ -0,0 +1,1125 @@
+{
+    "gold_tags": [
+        {
+            "object": [
+                "radiative transfer at second order"
+            ],
+            "method": [
+                "Boltzmann equation"
+            ]
+        },
+        {
+            "method": [
+                "Boltzmann equation"
+            ],
+            "object": [
+                "cosmological perturbations"
+            ],
+            "field": [
+                "cosmological"
+            ]
+        },
+        {
+            "object": [
+                "polarized radiation"
+            ],
+            "method": [
+                "tensor valued distribution function",
+                "distribution function",
+                "gauge-invariant distribution function"
+            ],
+            "property": [
+                "gauge dependence",
+                "gauge-invariant"
+            ]
+        },
+        {
+            "model": [
+                "Liouville operator"
+            ],
+            "object": [
+                "electrons",
+                "collision term",
+                "scattering of photons",
+                "free electrons"
+            ]
+        },
+        {
+            "property": [
+                "dependence",
+                "momentum",
+                "symmetric",
+                "trace-free"
+            ],
+            "object": [
+                "photon",
+                "multipoles",
+                "normal modes components"
+            ]
+        },
+        {
+            "object": [
+                "cosmic microwave background bispectrum",
+                "second order perturbations",
+                "non-linear evolution"
+            ],
+            "property": [
+                "bispectrum",
+                "primordial non-Gaussianity",
+                "non-linear"
+            ]
+        },
+        {
+            "object": [
+                "Jupiter XO-5b"
+            ]
+        },
+        {
+            "dataset": [
+                "HATNet observations"
+            ],
+            "instrument": [
+                "HATNet"
+            ],
+            "object": [
+                "XO-5b"
+            ]
+        },
+        {
+            "property": [
+                "correlation",
+                "spectral bisector variations",
+                "orbital phase"
+            ]
+        },
+        {
+            "modality": [
+                "spectroscopic measurements"
+            ],
+            "property": [
+                "scatter",
+                "spectral line bisectors"
+            ]
+        },
+        {
+            "object": [
+                "stellar configurations",
+                "hierarchical triples",
+                "chance alignments",
+                "planet signals"
+            ]
+        },
+        {
+            "property": [
+                "S activity index"
+            ],
+            "object": [
+                "stellar activity"
+            ]
+        },
+        {
+            "object": [
+                "planet parameters"
+            ]
+        },
+        {
+            "object": [
+                "stellar and planetary parameters"
+            ]
+        },
+        {
+            "object": [
+                "XO-5b",
+                "G type star"
+            ],
+            "property": [
+                "mass",
+                "M_s",
+                "radius",
+                "R_s",
+                "metallicity"
+            ]
+        },
+        {
+            "property": [
+                "mass",
+                "radius",
+                "M_p",
+                "R_p",
+                "mean density"
+            ]
+        },
+        {
+            "property": [
+                "ephemeris",
+                "orbit",
+                "P",
+                "E",
+                "transit duration"
+            ]
+        },
+        {
+            "object": [
+                "transit timing variations"
+            ]
+        },
+        {
+            "object": [
+                "XO-5b",
+                "exoplanets"
+            ],
+            "property": [
+                "Safronov number",
+                "gravity"
+            ]
+        },
+        {
+            "property": [
+                "Electron Temperatures",
+                "High Resolution",
+                "Low Frequency"
+            ],
+            "object": [
+                "W51 Complex"
+            ],
+            "modality": [
+                "Radio"
+            ]
+        },
+        {
+            "object": [
+                "W51",
+                "radio complex",
+                "Sagitarius arm",
+                "Sun"
+            ],
+            "modality": [
+                "radio"
+            ]
+        },
+        {
+            "object": [
+                "W51A",
+                "W51B",
+                "HII regions",
+                "W51C",
+                "supernova remnant"
+            ]
+        },
+        {
+            "dataset": [
+                "continuum radio observations"
+            ],
+            "modality": [
+                "continuum radio"
+            ],
+            "object": [
+                "HII regions",
+                "W51"
+            ],
+            "instrument": [
+                "GMRT"
+            ]
+        },
+        {
+            "modality": [
+                "spectra"
+            ],
+            "property": [
+                "thermal subcomponents",
+                "electron temperatures",
+                "emission measures"
+            ],
+            "object": [
+                "W51",
+                "emission spectrum"
+            ]
+        },
+        {
+            "property": [
+                "electron temperatures",
+                "temperatures"
+            ],
+            "modality": [
+                "continuum spectra"
+            ],
+            "method": [
+                "radio recombination line (RRL) studies"
+            ],
+            "object": [
+                "HII regions"
+            ]
+        },
+        {
+            "property": [
+                "brightness"
+            ],
+            "model": [
+                "multicomponent model"
+            ]
+        },
+        {
+            "object": [
+                "A1689",
+                "Wide-Field VLT/VIMOS Spectroscopy"
+            ],
+            "instrument": [
+                "VLT/VIMOS"
+            ],
+            "modality": [
+                "Spectroscopy"
+            ],
+            "property": [
+                "Mass Profile",
+                "Concentration Parameter",
+                "Velocity Anisotropy"
+            ]
+        },
+        {
+            "object": [
+                "cluster A1689"
+            ],
+            "modality": [
+                "VLT/VIMOS spectroscopy",
+                "Subaru/Suprime-Cam imaging"
+            ],
+            "instrument": [
+                "VLT/VIMOS",
+                "Subaru/Suprime-Cam"
+            ]
+        },
+        {
+            "property": [
+                "radial velocity distribution",
+                "amplitude",
+                "radius"
+            ],
+            "object": [
+                "cluster members"
+            ]
+        },
+        {
+            "property": [
+                "3D velocity anisotropy",
+                "galaxy number density",
+                "velocity dispersion",
+                "galaxy counts",
+                "cluster mass"
+            ],
+            "method": [
+                "Jeans equation"
+            ],
+            "instrument": [
+                "Subaru imaging"
+            ],
+            "modality": [
+                "joint lensing",
+                "X-ray"
+            ]
+        },
+        {
+            "property": [
+                "velocity anisotropy",
+                "radial",
+                "radius"
+            ]
+        },
+        {
+            "object": [
+                "galaxy"
+            ],
+            "method": [
+                "Jeans equation",
+                "NFW form for the mass distribution"
+            ],
+            "property": [
+                "caustic amplitude",
+                "escape velocity"
+            ]
+        },
+        {
+            "object": [
+                "cluster"
+            ],
+            "property": [
+                "virial mass"
+            ],
+            "modality": [
+                "lensing",
+                "X-ray"
+            ]
+        },
+        {
+            "property": [
+                "NFW concentration parameter"
+            ]
+        },
+        {
+            "property": [
+                "virial radius",
+                "radius",
+                "caustics",
+                "velocity"
+            ],
+            "object": [
+                "cluster",
+                "A1689"
+            ]
+        },
+        {
+            "property": [
+                "neutronization",
+                "core-collapse"
+            ],
+            "object": [
+                "supernovae"
+            ]
+        },
+        {
+            "property": [
+                "temperature dependence",
+                "neutronization"
+            ],
+            "object": [
+                "nuclear symmetry energy",
+                "stellar core",
+                "neutrino",
+                "gravitational collapse"
+            ]
+        },
+        {
+            "method": [
+                "one-zone simulation"
+            ],
+            "property": [
+                "collapse",
+                "beta equilibrium"
+            ],
+            "object": [
+                "lepton fraction"
+            ]
+        },
+        {
+            "property": [
+                "strength",
+                "neutron-rich"
+            ],
+            "object": [
+                "electron capture",
+                "electron",
+                "nuclei",
+                "supernova"
+            ]
+        },
+        {
+            "property": [
+                "temperature dependence",
+                "dissociation energy",
+                "strength"
+            ],
+            "object": [
+                "symmetry energy",
+                "deleptonization",
+                "shock wave energetics",
+                "nuclear electron capture",
+                "electron"
+            ]
+        },
+        {
+            "method": [
+                "one-zone model"
+            ],
+            "task": [
+                "numerical simulations"
+            ],
+            "object": [
+                "supernova explosions"
+            ]
+        },
+        {
+            "object": [
+                "Hadrons",
+                "Kerr-Newman Black Holes"
+            ]
+        },
+        {
+            "property": [
+                "scale invariance",
+                "\\\"strong gravity\\\""
+            ],
+            "model": [
+                "Einstein field equations"
+            ],
+            "object": [
+                "hadrons",
+                "black holes",
+                "gravitational coupling factor"
+            ]
+        },
+        {
+            "task": [
+                "retrodicting"
+            ],
+            "property": [
+                "mass",
+                "angular momentum"
+            ],
+            "object": [
+                "proton",
+                "Kerr black hole"
+            ]
+        },
+        {
+            "property": [
+                "mass",
+                "radius"
+            ],
+            "object": [
+                "proton"
+            ],
+            "method": [
+                "geometrodynamics form",
+                "full Kerr-Newman solution"
+            ],
+            "model": [
+                "Einstein-Maxwell equations"
+            ]
+        },
+        {
+            "property": [
+                "radius"
+            ],
+            "object": [
+                "alpha particle"
+            ],
+            "task": [
+                "retrodictive test"
+            ]
+        },
+        {
+            "task": [
+                "retrodictive test",
+                "retrodicted"
+            ],
+            "modality": [
+                "subatomic particle mass spectrum"
+            ],
+            "method": [
+                "Kerr solution of General Relativity"
+            ]
+        },
+        {
+            "object": [
+                "particle masses"
+            ],
+            "method": [
+                "Kerr solution"
+            ],
+            "property": [
+                "n",
+                "M",
+                "Planck mass"
+            ]
+        },
+        {
+            "object": [
+                "retrodicted masses"
+            ]
+        },
+        {
+            "object": [
+                "gravitational coupling constant",
+                "Planck scale",
+                "fine structure constant"
+            ]
+        },
+        {
+            "task": [
+                "Large-Scale Magnetic-Field Generation"
+            ],
+            "method": [
+                "Randomly Forced Shearing Waves"
+            ]
+        },
+        {
+            "task": [
+                "generation of a large-scale magnetic field"
+            ],
+            "object": [
+                "magnetic field",
+                "conducting fluid"
+            ],
+            "method": [
+                "random non-helically forced motions",
+                "linear shear"
+            ]
+        },
+        {
+            "property": [
+                "dynamo",
+                "kinematic"
+            ],
+            "object": [
+                "electromotive force"
+            ]
+        },
+        {
+            "property": [
+                "shear dynamo"
+            ]
+        },
+        {
+            "property": [
+                "wavenumber",
+                "growth rate"
+            ],
+            "object": [
+                "mode"
+            ]
+        },
+        {
+            "property": [
+                "shear dynamo action"
+            ],
+            "object": [
+                "magnetohydrodynamic turbulence"
+            ]
+        },
+        {
+            "object": [
+                "Sunyaev-Zel'dovich Effect",
+                "Bullet Cluster (1E 0657-56)"
+            ],
+            "instrument": [
+                "APEX-SZ"
+            ]
+        },
+        {
+            "object": [
+                "Sunyaev-Zel'dovich effect (SZE)",
+                "Bullet cluster (1E 0657--56)"
+            ],
+            "instrument": [
+                "APEX-SZ"
+            ]
+        },
+        {
+            "object": [
+                "SZE",
+                "galaxy cluster"
+            ]
+        },
+        {
+            "object": [
+                "cluster"
+            ],
+            "property": [
+                "radius"
+            ]
+        },
+        {
+            "object": [
+                "SZE map",
+                "X-ray maps",
+                "SZE emission"
+            ],
+            "modality": [
+                "X-ray",
+                "radio",
+                "IR"
+            ]
+        },
+        {
+            "model": [
+                "cluster gas models"
+            ],
+            "object": [
+                "cluster gas"
+            ],
+            "dataset": [
+                "X-ray observations"
+            ],
+            "modality": [
+                "X-ray"
+            ],
+            "method": [
+                "isothermal elliptical beta model"
+            ]
+        },
+        {
+            "modality": [
+                "X-ray"
+            ],
+            "object": [
+                "power-law index"
+            ],
+            "property": [
+                "radius",
+                "axial ratio",
+                "temperature decrement",
+                "flux calibration uncertainty"
+            ]
+        },
+        {
+            "object": [
+                "APEX-SZ map",
+                "cluster gas"
+            ],
+            "instrument": [
+                "APEX-SZ",
+                "Chandra X-ray observations",
+                "X-ray observations"
+            ],
+            "dataset": [
+                "Chandra X-ray observations"
+            ],
+            "modality": [
+                "X-ray",
+                "X-ray spectroscopic"
+            ],
+            "property": [
+                "mass-weighted",
+                "temperature"
+            ]
+        },
+        {
+            "object": [
+                "cluster gas",
+                "prolate",
+                "oblate spheroidal geometries"
+            ],
+            "property": [
+                "gas mass fraction"
+            ],
+            "modality": [
+                "X-ray",
+                "weak lensing"
+            ]
+        },
+        {
+            "instrument": [
+                "APEX-SZ",
+                "multiplexed superconducting transition-edge sensor bolometers"
+            ]
+        },
+        {
+            "model": [
+                "Natural Coordinate System"
+            ],
+            "object": [
+                "Curved Space-time"
+            ]
+        },
+        {
+            "model": [
+                "coordinate system"
+            ],
+            "object": [
+                "curved space-time",
+                "hypersurface orthogonal",
+                "time coordinate"
+            ]
+        },
+        {
+            "object": [
+                "time coordinate"
+            ]
+        },
+        {
+            "model": [
+                "coordinate system"
+            ],
+            "property": [
+                "geometrical"
+            ]
+        },
+        {
+            "model": [
+                "\\\"natural coordinate system\\\""
+            ]
+        },
+        {
+            "model": [
+                "natural coordinate system"
+            ],
+            "property": [
+                "physcial",
+                "geometrical"
+            ]
+        },
+        {
+            "model": [
+                "NCS"
+            ],
+            "method": [
+                "classical approximation of spinor equation",
+                "weak field approximation of Einstein's equation",
+                "low speed approximation of particles"
+            ],
+            "object": [
+                "Newtonian mechanics",
+                "space-time"
+            ]
+        },
+        {
+            "model": [
+                "coordinate system"
+            ],
+            "object": [
+                "curved space-time",
+                "space-time"
+            ]
+        },
+        {},
+        {
+            "method": [
+                "The Tully-Fisher Relation"
+            ],
+            "object": [
+                "the 2MASS J, H and K Bands"
+            ]
+        },
+        {
+            "dataset": [
+                "The 2 Micron All-Sky Survey (2MASS)",
+                "Tully-Fisher Survey (2MTF)",
+                "the 2MASS Redshift Survey (2MRS)"
+            ],
+            "property": [
+                "Tully-Fisher (TF) distances"
+            ],
+            "object": [
+                "bright inclined spirals"
+            ]
+        },
+        {
+            "method": [
+                "universal calibration",
+                "TF relation"
+            ],
+            "object": [
+                "the 2MASS J (1.2 um), H (1.6 um) and K (2.2 um) bands"
+            ]
+        },
+        {
+            "model": [
+                "bias corrected or universal TF template"
+            ]
+        },
+        {
+            "method": [
+                "TF relation"
+            ],
+            "property": [
+                "the wavelength"
+            ],
+            "object": [
+                "K-band",
+                "J and H-bands"
+            ]
+        },
+        {
+            "property": [
+                "galaxy morphology",
+                "TF zeropoint"
+            ],
+            "object": [
+                "later type spirals",
+                "earlier type spirals"
+            ]
+        },
+        {
+            "object": [
+                "Sc galaxies"
+            ]
+        },
+        {
+            "method": [
+                "TF relation"
+            ],
+            "object": [
+                "wavelength"
+            ]
+        },
+        {
+            "object": [
+                "S1 Shell",
+                "Interstellar Magnetic Field",
+                "the Heliosphere"
+            ]
+        },
+        {
+            "object": [
+                "Loop I magnetic superbubble",
+                "the Sun"
+            ]
+        },
+        {
+            "object": [
+                "the polarized radio continuum",
+                "Loop I",
+                "magnetic shells",
+                "the Sun",
+                "the 'S1' shell"
+            ]
+        },
+        {
+            "object": [
+                "the Sun",
+                "the local interstellar magnetic field",
+                "interstellar material"
+            ]
+        },
+        {
+            "modality": [
+                "optical polarization data"
+            ],
+            "object": [
+                "interstellar magnetic field (ISMF)",
+                "interstellar FeII and CaII",
+                "the Sun"
+            ]
+        },
+        {
+            "object": [
+                "Loop I",
+                "magnetic subshells",
+                "low density ISM"
+            ]
+        },
+        {
+            "property": [
+                "Column densities",
+                "star distance",
+                "total pathlength"
+            ],
+            "object": [
+                "the recombinant ion CaII",
+                "the interstellar radiation field",
+                "magnetic subshells"
+            ]
+        },
+        {
+            "property": [
+                "the dipole moment"
+            ],
+            "object": [
+                "the cosmic microwave background",
+                "the heliosphere",
+                "local ISMF"
+            ]
+        },
+        {
+            "object": [
+                "supermassive black holes",
+                "pseudo-bulges",
+                "classical bulges",
+                "elliptical galaxies"
+            ]
+        },
+        {
+            "method": [
+                "structural analysis"
+            ],
+            "object": [
+                "local galaxies",
+                "central black holes",
+                "elliptical galaxies",
+                "classical bulges",
+                "pseudo-bulges"
+            ],
+            "dataset": [
+                "Sloan Digital Sky Survey"
+            ],
+            "property": [
+                "mass",
+                "stellar masses",
+                "central stellar velocity dispersion sigma"
+            ]
+        },
+        {
+            "property": [
+                "elliptical galaxy/bulge mass",
+                "central black hole mass",
+                "mass"
+            ],
+            "object": [
+                "black holes",
+                "local universe",
+                "elliptical galaxies",
+                "classical bulges",
+                "pseudo-bulges"
+            ]
+        },
+        {
+            "object": [
+                "ellipticals",
+                "classical bulges",
+                "pseudo-bulges",
+                "barred galaxies"
+            ],
+            "property": [
+                "stellar masses",
+                "sigma"
+            ]
+        },
+        {
+            "property": [
+                "black hole masses"
+            ]
+        },
+        {
+            "property": [
+                "total mass density"
+            ],
+            "object": [
+                "black holes"
+            ]
+        },
+        {
+            "property": [
+                "Characteristic Scales"
+            ],
+            "object": [
+                "Initial Density and Velocity Fields"
+            ]
+        },
+        {
+            "object": [
+                "initial fields",
+                "density contrast",
+                "peculiar velocity",
+                "random field"
+            ],
+            "property": [
+                "differential and integral length scales"
+            ]
+        },
+        {
+            "property": [
+                "length scales",
+                "mass scales",
+                "length and mass scales"
+            ],
+            "object": [
+                "halos of young galaxies",
+                "halos of galaxies",
+                "galaxy distribution"
+            ]
+        },
+        {
+            "property": [
+                "characteristic scales",
+                "initial fields"
+            ]
+        },
+        {
+            "object": [
+                "Gravitino Dark Matter",
+                "R-violating Operators"
+            ]
+        },
+        {
+            "object": [
+                "gravitino dark matter",
+                "gravitino decays",
+                "R-violating supersymmetry",
+                "R-violating couplings"
+            ],
+            "property": [
+                "flavour dependence"
+            ]
+        },
+        {
+            "property": [
+                "decay modes",
+                "final state products"
+            ]
+        },
+        {
+            "method": [
+                "Mixing effects"
+            ],
+            "property": [
+                "magnitude",
+                "phase space suppression"
+            ]
+        },
+        {
+            "property": [
+                "strength of different decay rates"
+            ],
+            "object": [
+                "gravitino",
+                "Next-to-Lightest Supersymmetric Particle",
+                "superparticle"
+            ]
+        },
+        {
+            "object": [
+                "Dark Matter Caustics",
+                "Galaxy Clusters"
+            ]
+        },
+        {
+            "modality": [
+                "gravitational lensing observations"
+            ],
+            "object": [
+                "gravitational lensing"
+            ]
+        },
+        {
+            "object": [
+                "{\\\\it caustic} ring",
+                "dark matter",
+                "galaxy cluster"
+            ]
+        },
+        {
+            "object": [
+                "caustic ring",
+                "flow",
+                "gravitational potential well"
+            ],
+            "property": [
+                "cold",
+                "collisionless",
+                "net overall rotation"
+            ]
+        },
+        {
+            "object": [
+                "caustic rings",
+                "dark matter",
+                "Milky Way",
+                "galaxies"
+            ],
+            "property": [
+                "isolated",
+                "spiral"
+            ]
+        },
+        {
+            "object": [
+                "galaxy clusters",
+                "caustic rings"
+            ]
+        },
+        {
+            "task": [
+                "calculate the column density profile",
+                "show that it is consistent with the observations of"
+            ],
+            "property": [
+                "column density profile"
+            ],
+            "object": [
+                "caustic ring",
+                "cluster"
+            ]
+        },
+        "property",
+        "object"
+    ]
+}

data/results/eval_random_readable_zoOLWDdW_20240716_040746/metrics.json

@@ -0,0 +1,13 @@
+{
+    "metrics": {
+        "avg_pred_response_time_per_sentence": 2.7855,
+        "total_time": 373.287,
+        "precision": 0.407,
+        "recall": 0.2949,
+        "f1": 0.342,
+        "union_precision": 0.5865,
+        "union_recall": 0.4247,
+        "union_f1": 0.4927
+    },
+    "prompt_file": "prompts.txt"
+}

data/results/eval_random_readable_zoOLWDdW_20240716_040746/predicted_responses.txt

@@ -0,0 +1,1119 @@
+assistant
+
+method: radiative transfer
+##################################################
+assistant
+
+model: Boltzmann equation
+##################################################
+assistant
+
+model: tensor valued distribution function, gauge-invariant distribution function
+object: polarized radiation, distribution function
+##################################################
+assistant
+
+method: Liouville operator, collision term
+##################################################
+assistant
+
+modality: multipoles, normal modes
+##################################################
+assistant
+
+model: cosmic microwave background
+object: perturbations, non-Gaussianity, evolution
+##################################################
+assistant
+
+object: XO-5b, Jupiter
+##################################################
+assistant
+
+dataset: HATNet observations
+object: XO-5b, planetary nature
+##################################################
+assistant
+
+property: spectral bisector variations, orbital phase
+##################################################
+assistant
+
+dataset: spectroscopic measurements
+##################################################
+assistant
+
+Extractions:
+object: planet signals, data
+##################################################
+assistant
+
+object: S activity index, stellar activity
+##################################################
+assistant
+
+object: planet parameters
+##################################################
+assistant
+
+No extractions.
+##################################################
+assistant
+
+object: XO-5b, star
+property: mass, radius, metallicity
+##################################################
+assistant
+
+property: planetary mass, radius, mean density
+object: planet
+##################################################
+assistant
+
+property: P, E, transit duration
+##################################################
+assistant
+
+property: transit timing variations
+##################################################
+assistant
+
+object: XO-5b, exoplanets
+property: Safronov number, surface gravity
+##################################################
+assistant
+
+object: Electron Temperatures, W51 Complex
+##################################################
+assistant
+
+object: W51, Sagitarius arm, Sun 
+property: distance, extension
+##################################################
+assistant
+
+object: W51A, W51B, W51C, HII regions, supernova remnant
+##################################################
+assistant
+
+dataset: continuum radio observations
+instrument: GMRT
+modality: radio
+object: HII regions, W51 complex
+##################################################
+assistant
+
+object: W51, thermal subcomponents
+property: electron temperatures, emission measures
+modality: spectra
+##################################################
+assistant
+
+instrument: continuum spectra, radio recombination line (RRL)
+modality: spectra
+property: electron temperatures, filling factor
+##################################################
+assistant
+
+model: multicomponent model
+##################################################
+assistant
+
+object: A1689, Mass Profile, Concentration Parameter, Velocity Anisotropy
+##################################################
+assistant
+
+dataset: VLT/VIMOS spectroscopy, Subaru/Suprime-Cam imaging
+instrument: VLT/VIMOS, Subaru/Suprime-Cam
+object: cluster A1689
+##################################################
+assistant
+
+property: radial velocity distribution, amplitude, velocity, radius
+object: cluster members, velocity caustics
+##################################################
+assistant
+
+model: Jeans equation
+method: model-independent method
+object: velocity anisotropy, galaxy, velocity dispersion, galaxy counts, cluster mass profile
+property: 3D
+##################################################
+assistant
+
+property: velocity anisotropy
+##################################################
+assistant
+
+method: solution of the Jeans equation, method
+object: galaxy, mass distribution, caustic amplitude, escape velocity
+##################################################
+assistant
+
+model: dynamical methods
+object: cluster virial mass, lensing, X-ray analysis
+##################################################
+assistant
+
+property: NFW concentration parameter
+##################################################
+assistant
+
+object: virial radius, caustics, cluster members, A1689
+property: velocity
+##################################################
+assistant
+
+object: neutronization, core-collapse supernovae
+##################################################
+assistant
+
+property: temperature dependence, nuclear symmetry energy
+object: stellar core
+##################################################
+assistant
+
+model: one-zone simulation
+task: follow the collapse
+##################################################
+assistant
+
+Extractions:
+property: strength
+##################################################
+assistant
+
+property: temperature dependence, symmetry energy, dissociation energy, shock wave energetics
+##################################################
+assistant
+
+model: one-zone model
+##################################################
+assistant
+
+object: Hadrons, Kerr-Newman Black Holes
+##################################################
+assistant
+
+model: Einstein field equations
+##################################################
+assistant
+
+model: hypothesis
+object: proton, Kerr black hole
+property: mass, angular momentum
+##################################################
+assistant
+
+model: Kerr-Newman solution of the Einstein-Maxwell equations
+object: proton
+##################################################
+assistant
+
+object: alpha particle
+property: radius
+##################################################
+assistant
+
+model: Kerr solution of General Relativity
+object: subatomic particle mass spectrum
+##################################################
+assistant
+
+property: particle masses, Kerr solution, Planck mass
+object: particle
+##################################################
+assistant
+
+property: accuracy
+##################################################
+assistant
+
+property: fine structure constant
+##################################################
+assistant
+
+model: Large-Scale Magnetic-Field Generation
+##################################################
+assistant
+
+model: rigorous theory
+method: random non-helically forced motions
+property: large-scale, low Rm, weak shear
+object: magnetic field, conducting fluid
+##################################################
+assistant
+
+object: dynamo, electromotive force
+##################################################
+assistant
+
+model: quasilinear calculation, shear dynamo
+##################################################
+assistant
+
+method: analytically
+##################################################
+assistant
+
+model: shear dynamo action
+##################################################
+assistant
+
+object: Bullet Cluster, Sunyaev-Zel'dovich Effect
+##################################################
+assistant
+
+instrument: APEX-SZ
+object: Sunyaev-Zel'dovich effect (SZE), Bullet cluster
+##################################################
+assistant
+
+object: SZE, galaxy cluster
+##################################################
+assistant
+
+property: significance
+##################################################
+assistant
+
+object: SZE map, X-ray maps, SZE emission, radio sources, IR sources
+##################################################
+assistant
+
+model: isothermal elliptical beta model
+##################################################
+assistant
+
+property: power-law index, beta, core radius, axial ratio, central temperature decrement, flux calibration uncertainty
+##################################################
+assistant
+
+dataset: APEX-SZ map, Chandra X-ray observations
+instrument: APEX-SZ, Chandra
+property: mass-weighted temperature, X-ray spectroscopic temperatures
+##################################################
+assistant
+
+model: isothermal cluster gas distribution
+object: gas, cluster, X-ray
+##################################################
+assistant
+
+instrument: APEX-SZ experiment
+##################################################
+assistant
+
+field: Curved Space-time
+##################################################
+assistant
+
+field: space-time
+##################################################
+assistant
+
+Extractions:
+property: time
+##################################################
+assistant
+
+Extractions:
+##################################################
+assistant
+
+No extractions.
+##################################################
+assistant
+
+method: constructing method
+##################################################
+assistant
+
+model: classical approximation of spinor equation, Einstein's equation
+method: classical approximation, weak field approximation, low speed approximation
+##################################################
+assistant
+
+model: coordinate system
+##################################################
+assistant
+
+field:
+##################################################
+assistant
+
+object: Tully-Fisher Relation
+##################################################
+assistant
+
+dataset: 2 Micron All-Sky Survey (2MASS), 2MASS Redshift Survey (2MRS)
+instrument: 2 Micron All-Sky Survey (2MASS)
+object: Tully-Fisher distances, spirals
+##################################################
+assistant
+
+method: TF relation
+##################################################
+assistant
+
+model: TF template
+##################################################
+assistant
+
+property: slope, wavelength
+object: TF relation, K-band, J, H-bands
+##################################################
+assistant
+
+object: galaxy, spirals
+property: morphology
+##################################################
+assistant
+
+field: Sc galaxies
+##################################################
+assistant
+
+property: width, intrinsic scatter
+##################################################
+assistant
+
+object: S1 Shell, Interstellar Magnetic Field, Gas, Heliosphere
+##################################################
+assistant
+
+object: Loop I magnetic superbubble, Sun
+##################################################
+assistant
+
+model: magnetic shells
+object: Loop I, Sun, 'S1' shell
+##################################################
+assistant
+
+object: Sun, interstellar magnetic field, interstellar material
+##################################################
+assistant
+
+object: subshells, interstellar magnetic field (ISMF), interstellar FeII, CaII, Sun
+property: properties
+##################################################
+assistant
+
+model: Loop I
+object: ISM
+##################################################
+assistant
+
+object: CaII, interstellar radiation field, star, magnetic subshells
+property: column densities, strength
+##################################################
+assistant
+
+object: dipole moment, cosmic microwave background, heliosphere, local ISMF
+##################################################
+assistant
+
+object: supermassive black holes, pseudo-bulges, classical bulges, elliptical galaxies
+##################################################
+assistant
+
+dataset: Sloan Digital Sky Survey
+object: galaxies, elliptical galaxies, classical bulges, pseudo-bulges, black holes
+property: mass, stellar masses, central stellar velocity dispersion
+##################################################
+assistant
+
+object: elliptical galaxy, bulge, black hole, local universe, classical bulges, pseudo-bulges
+property: mass
+##################################################
+assistant
+
+object: ellipticals, classical bulges, pseudo-bulges, stellar masses, sigma, barred galaxies
+##################################################
+assistant
+
+object: black hole, M_BH-M_Bulge, M_BH-sigma relations
+##################################################
+assistant
+
+property: total mass density
+##################################################
+assistant
+
+property: Characteristic Scales, Initial Density, Velocity Fields
+##################################################
+assistant
+
+property: density contrast, peculiar velocity, differential length scales, integral length scales
+object: field
+##################################################
+assistant
+
+object: halos, galaxies, galaxy distribution
+##################################################
+assistant
+
+object: scales, fields
+##################################################
+assistant
+
+object: Gravitino Dark Matter, Flavour Structure, R-violating Operators
+##################################################
+assistant
+
+model: R-violating supersymmetry
+object: gravitino dark matter, gravitino
+##################################################
+assistant
+
+property: dominant decay modes, R-violating hierarchies
+##################################################
+assistant
+
+property: relative magnitude, phase space suppression
+##################################################
+assistant
+
+object: gravitino, Next-to-Lightest Supersymmetric Particle (NLSP), superparticle
+##################################################
+assistant
+
+object: Dark Matter Caustics, Galaxy Clusters
+##################################################
+assistant
+
+object: gravitational lensing observations, Jee
+##################################################
+assistant
+
+object: caustic ring, dark matter, galaxy cluster
+##################################################
+assistant
+
+model: gravitational potential well
+##################################################
+assistant
+
+object: caustic rings, dark matter, Milky Way, spiral galaxies
+##################################################
+assistant
+
+object: galaxy clusters, caustic rings
+##################################################
+assistant
+
+model: 
+task: 
+dataset: 
+field: 
+modality: 
+method: 
+object: caustic ring, cluster
+property: column density profile
+##################################################
+assistant
+
+method: radiative transfer
+##################################################
+assistant
+
+model: Boltzmann equation
+##################################################
+assistant
+
+model: tensor valued distribution function, gauge-invariant distribution function
+object: polarized radiation, distribution function
+##################################################
+assistant
+
+method: Liouville operator, collision term
+##################################################
+assistant
+
+modality: multipoles, normal modes
+##################################################
+assistant
+
+model: cosmic microwave background
+object: perturbations, non-Gaussianity, evolution
+##################################################
+assistant
+
+object: XO-5b, Jupiter
+##################################################
+assistant
+
+dataset: HATNet observations
+object: XO-5b, planetary nature
+##################################################
+assistant
+
+property: spectral bisector variations, orbital phase
+##################################################
+assistant
+
+dataset: spectroscopic measurements
+##################################################
+assistant
+
+Extractions:
+object: planet signals, data
+##################################################
+assistant
+
+object: S activity index, stellar activity
+##################################################
+assistant
+
+object: planet parameters
+##################################################
+assistant
+
+No extractions.
+##################################################
+assistant
+
+object: XO-5b, star
+property: mass, radius, metallicity
+##################################################
+assistant
+
+property: planetary mass, radius, mean density
+object: planet
+##################################################
+assistant
+
+property: P, E, transit duration
+##################################################
+assistant
+
+property: transit timing variations
+##################################################
+assistant
+
+object: XO-5b, exoplanets
+property: Safronov number, surface gravity
+##################################################
+assistant
+
+object: Electron Temperatures, W51 Complex
+##################################################
+assistant
+
+object: W51, Sagitarius arm, Sun 
+property: distance, extension
+##################################################
+assistant
+
+object: W51A, W51B, W51C, HII regions, supernova remnant
+##################################################
+assistant
+
+dataset: continuum radio observations
+instrument: GMRT
+modality: radio
+object: HII regions, W51 complex
+##################################################
+assistant
+
+object: W51, thermal subcomponents
+property: electron temperatures, emission measures
+modality: spectra
+##################################################
+assistant
+
+instrument: continuum spectra, radio recombination line (RRL)
+modality: spectra
+property: electron temperatures, filling factor
+##################################################
+assistant
+
+model: multicomponent model
+##################################################
+assistant
+
+object: A1689, Mass Profile, Concentration Parameter, Velocity Anisotropy
+##################################################
+assistant
+
+dataset: VLT/VIMOS spectroscopy, Subaru/Suprime-Cam imaging
+instrument: VLT/VIMOS, Subaru/Suprime-Cam
+object: cluster A1689
+##################################################
+assistant
+
+property: radial velocity distribution, amplitude, velocity, radius
+object: cluster members, velocity caustics
+##################################################
+assistant
+
+model: Jeans equation
+method: model-independent method
+object: velocity anisotropy, galaxy, velocity dispersion, galaxy counts, cluster mass profile
+property: 3D
+##################################################
+assistant
+
+property: velocity anisotropy
+##################################################
+assistant
+
+method: solution of the Jeans equation, method
+object: galaxy, mass distribution, caustic amplitude, escape velocity
+##################################################
+assistant
+
+model: dynamical methods
+object: cluster virial mass, lensing, X-ray analysis
+##################################################
+assistant
+
+property: NFW concentration parameter
+##################################################
+assistant
+
+object: virial radius, caustics, cluster members, A1689
+property: velocity
+##################################################
+assistant
+
+object: neutronization, core-collapse supernovae
+##################################################
+assistant
+
+property: temperature dependence, nuclear symmetry energy
+object: stellar core
+##################################################
+assistant
+
+model: one-zone simulation
+task: follow the collapse
+##################################################
+assistant
+
+Extractions:
+property: strength
+##################################################
+assistant
+
+property: temperature dependence, symmetry energy, dissociation energy, shock wave energetics
+##################################################
+assistant
+
+model: one-zone model
+##################################################
+assistant
+
+object: Hadrons, Kerr-Newman Black Holes
+##################################################
+assistant
+
+model: Einstein field equations
+##################################################
+assistant
+
+model: hypothesis
+object: proton, Kerr black hole
+property: mass, angular momentum
+##################################################
+assistant
+
+model: Kerr-Newman solution of the Einstein-Maxwell equations
+object: proton
+##################################################
+assistant
+
+object: alpha particle
+property: radius
+##################################################
+assistant
+
+model: Kerr solution of General Relativity
+object: subatomic particle mass spectrum
+##################################################
+assistant
+
+property: particle masses, Kerr solution, Planck mass
+object: particle
+##################################################
+assistant
+
+property: accuracy
+##################################################
+assistant
+
+property: fine structure constant
+##################################################
+assistant
+
+model: Large-Scale Magnetic-Field Generation
+##################################################
+assistant
+
+model: rigorous theory
+method: random non-helically forced motions
+property: large-scale, low Rm, weak shear
+object: magnetic field, conducting fluid
+##################################################
+assistant
+
+object: dynamo, electromotive force
+##################################################
+assistant
+
+model: quasilinear calculation, shear dynamo
+##################################################
+assistant
+
+method: analytically
+##################################################
+assistant
+
+model: shear dynamo action
+##################################################
+assistant
+
+object: Bullet Cluster, Sunyaev-Zel'dovich Effect
+##################################################
+assistant
+
+instrument: APEX-SZ
+object: Sunyaev-Zel'dovich effect (SZE), Bullet cluster
+##################################################
+assistant
+
+object: SZE, galaxy cluster
+##################################################
+assistant
+
+property: significance
+##################################################
+assistant
+
+object: SZE map, X-ray maps, SZE emission, radio sources, IR sources
+##################################################
+assistant
+
+model: isothermal elliptical beta model
+##################################################
+assistant
+
+property: power-law index, beta, core radius, axial ratio, central temperature decrement, flux calibration uncertainty
+##################################################
+assistant
+
+dataset: APEX-SZ map, Chandra X-ray observations
+instrument: APEX-SZ, Chandra
+property: mass-weighted temperature, X-ray spectroscopic temperatures
+##################################################
+assistant
+
+model: isothermal cluster gas distribution
+object: gas, cluster, X-ray
+##################################################
+assistant
+
+instrument: APEX-SZ experiment
+##################################################
+assistant
+
+field: Curved Space-time
+##################################################
+assistant
+
+field: space-time
+##################################################
+assistant
+
+Extractions:
+property: time
+##################################################
+assistant
+
+Extractions:
+##################################################
+assistant
+
+No extractions.
+##################################################
+assistant
+
+method: constructing method
+##################################################
+assistant
+
+model: classical approximation of spinor equation, Einstein's equation
+method: classical approximation, weak field approximation, low speed approximation
+##################################################
+assistant
+
+model: coordinate system
+##################################################
+assistant
+
+field:
+##################################################
+assistant
+
+object: Tully-Fisher Relation
+##################################################
+assistant
+
+dataset: 2 Micron All-Sky Survey (2MASS), 2MASS Redshift Survey (2MRS)
+instrument: 2 Micron All-Sky Survey (2MASS)
+object: Tully-Fisher distances, spirals
+##################################################
+assistant
+
+method: TF relation
+##################################################
+assistant
+
+model: TF template
+##################################################
+assistant
+
+property: slope, wavelength
+object: TF relation, K-band, J, H-bands
+##################################################
+assistant
+
+object: galaxy, spirals
+property: morphology
+##################################################
+assistant
+
+field: Sc galaxies
+##################################################
+assistant
+
+property: width, intrinsic scatter
+##################################################
+assistant
+
+object: S1 Shell, Interstellar Magnetic Field, Gas, Heliosphere
+##################################################
+assistant
+
+object: Loop I magnetic superbubble, Sun
+##################################################
+assistant
+
+model: magnetic shells
+object: Loop I, Sun, 'S1' shell
+##################################################
+assistant
+
+object: Sun, interstellar magnetic field, interstellar material
+##################################################
+assistant
+
+object: subshells, interstellar magnetic field (ISMF), interstellar FeII, CaII, Sun
+property: properties
+##################################################
+assistant
+
+model: Loop I
+object: ISM
+##################################################
+assistant
+
+object: CaII, interstellar radiation field, star, magnetic subshells
+property: column densities, strength
+##################################################
+assistant
+
+object: dipole moment, cosmic microwave background, heliosphere, local ISMF
+##################################################
+assistant
+
+object: supermassive black holes, pseudo-bulges, classical bulges, elliptical galaxies
+##################################################
+assistant
+
+dataset: Sloan Digital Sky Survey
+object: galaxies, elliptical galaxies, classical bulges, pseudo-bulges, black holes
+property: mass, stellar masses, central stellar velocity dispersion
+##################################################
+assistant
+
+object: elliptical galaxy, bulge, black hole, local universe, classical bulges, pseudo-bulges
+property: mass
+##################################################
+assistant
+
+object: ellipticals, classical bulges, pseudo-bulges, stellar masses, sigma, barred galaxies
+##################################################
+assistant
+
+object: black hole, M_BH-M_Bulge, M_BH-sigma relations
+##################################################
+assistant
+
+property: total mass density
+##################################################
+assistant
+
+property: Characteristic Scales, Initial Density, Velocity Fields
+##################################################
+assistant
+
+property: density contrast, peculiar velocity, differential length scales, integral length scales
+object: field
+##################################################
+assistant
+
+object: halos, galaxies, galaxy distribution
+##################################################
+assistant
+
+object: scales, fields
+##################################################
+assistant
+
+object: Gravitino Dark Matter, Flavour Structure, R-violating Operators
+##################################################
+assistant
+
+model: R-violating supersymmetry
+object: gravitino dark matter, gravitino
+##################################################
+assistant
+
+property: dominant decay modes, R-violating hierarchies
+##################################################
+assistant
+
+property: relative magnitude, phase space suppression
+##################################################
+assistant
+
+object: gravitino, Next-to-Lightest Supersymmetric Particle (NLSP), superparticle
+##################################################
+assistant
+
+object: Dark Matter Caustics, Galaxy Clusters
+##################################################
+assistant
+
+object: gravitational lensing observations, Jee
+##################################################
+assistant
+
+object: caustic ring, dark matter, galaxy cluster
+##################################################
+assistant
+
+model: gravitational potential well
+##################################################
+assistant
+
+object: caustic rings, dark matter, Milky Way, spiral galaxies
+##################################################
+assistant
+
+object: galaxy clusters, caustic rings
+##################################################
+assistant
+
+model: 
+task: 
+dataset: 
+field: 
+modality: 
+method: 
+object: caustic ring, cluster
+property: column density profile
+##################################################
+assistant
+
+field: dark matter
+##################################################
+assistant
+
+task: estimating the mass
+##################################################
+assistant
+
+model: 4-component model
+object: elliptical galaxies, stars, dark matter (DM), hot gas, central black hole
+##################################################
+assistant
+
+object: DM, stars, central black hole
+##################################################
+assistant
+
+model: DM density profile
+##################################################
+assistant
+
+property: gas fraction, baryon fraction
+object: radius, effective radii (R_e)
+##################################################
+assistant
+
+property: line-of-sight velocity dispersion, velocity anisotropy, total mass
+object: velocity
+##################################################
+assistant
+
+model: LCDM simulations
+object: DM distributions, stellar VD profiles, VDs
+##################################################
+assistant
+
+object: planetary nebulae, baryon fraction
+##################################################
+assistant
+
+model: NFW DM model, Navarro et al
+##################################################
+assistant
+
+Extractions:
+##################################################
+assistant
+
+model: NFW model
+##################################################
+assistant
+
+model: LCDM, NFW
+property: radial anisotropy, M/L
+##################################################
+assistant
+
+model: DM model
+##################################################
+assistant
+
+model: anisotropic models, Osipkov-Merritt model
+##################################################
+assistant
+
+object: hyperons, neutron stars
+##################################################
+assistant
+
+model: equation of state
+object: hyperon-nucleon potential, nucleonic matter
+##################################################
+assistant
+
+method: renormalization group framework
+object: hyperon-nucleon potential
+##################################################
+assistant
+
+model: Hartree-Fock approximation
+##################################################
+assistant
+
+object: hyperons, neutron stars
+##################################################
+assistant
+
+object: neutron stars
+##################################################

data/results/eval_random_readable_zoOLWDdW_20240716_040746/predictions.json

@@ -0,0 +1,1886 @@
+{
+    "predicted_tags": [
+        {
+            "method": [
+                "radiative transfer"
+            ]
+        },
+        {
+            "model": [
+                "Boltzmann equation"
+            ]
+        },
+        {
+            "model": [
+                "tensor valued distribution function",
+                "gauge-invariant distribution function"
+            ],
+            "object": [
+                "polarized radiation",
+                "distribution function"
+            ]
+        },
+        {
+            "method": [
+                "Liouville operator",
+                "collision term"
+            ]
+        },
+        {
+            "modality": [
+                "multipoles",
+                "normal modes"
+            ]
+        },
+        {
+            "model": [
+                "cosmic microwave background"
+            ],
+            "object": [
+                "perturbations",
+                "non-Gaussianity",
+                "evolution"
+            ]
+        },
+        {
+            "object": [
+                "XO-5b",
+                "Jupiter"
+            ]
+        },
+        {
+            "dataset": [
+                "HATNet observations"
+            ],
+            "object": [
+                "XO-5b",
+                "planetary nature"
+            ]
+        },
+        {
+            "property": [
+                "spectral bisector variations",
+                "orbital phase"
+            ]
+        },
+        {
+            "dataset": [
+                "spectroscopic measurements"
+            ]
+        },
+        {
+            "object": [
+                "planet signals",
+                "data"
+            ]
+        },
+        {
+            "object": [
+                "S activity index",
+                "stellar activity"
+            ]
+        },
+        {
+            "object": [
+                "planet parameters"
+            ]
+        },
+        {},
+        {
+            "object": [
+                "XO-5b",
+                "star"
+            ],
+            "property": [
+                "mass",
+                "radius",
+                "metallicity"
+            ]
+        },
+        {
+            "property": [
+                "planetary mass",
+                "radius",
+                "mean density"
+            ],
+            "object": [
+                "planet"
+            ]
+        },
+        {
+            "property": [
+                "P",
+                "E",
+                "transit duration"
+            ]
+        },
+        {
+            "property": [
+                "transit timing variations"
+            ]
+        },
+        {
+            "object": [
+                "XO-5b",
+                "exoplanets"
+            ],
+            "property": [
+                "Safronov number",
+                "surface gravity"
+            ]
+        },
+        {
+            "object": [
+                "Electron Temperatures",
+                "W51 Complex"
+            ]
+        },
+        {
+            "object": [
+                "W51",
+                "Sagitarius arm",
+                "Sun "
+            ],
+            "property": [
+                "distance",
+                "extension"
+            ]
+        },
+        {
+            "object": [
+                "W51A",
+                "W51B",
+                "W51C",
+                "HII regions",
+                "supernova remnant"
+            ]
+        },
+        {
+            "dataset": [
+                "continuum radio observations"
+            ],
+            "instrument": [
+                "GMRT"
+            ],
+            "modality": [
+                "radio"
+            ],
+            "object": [
+                "HII regions",
+                "W51 complex"
+            ]
+        },
+        {
+            "object": [
+                "W51",
+                "thermal subcomponents"
+            ],
+            "property": [
+                "electron temperatures",
+                "emission measures"
+            ],
+            "modality": [
+                "spectra"
+            ]
+        },
+        {
+            "instrument": [
+                "continuum spectra",
+                "radio recombination line (RRL)"
+            ],
+            "modality": [
+                "spectra"
+            ],
+            "property": [
+                "electron temperatures",
+                "filling factor"
+            ]
+        },
+        {
+            "model": [
+                "multicomponent model"
+            ]
+        },
+        {
+            "object": [
+                "A1689",
+                "Mass Profile",
+                "Concentration Parameter",
+                "Velocity Anisotropy"
+            ]
+        },
+        {
+            "dataset": [
+                "VLT/VIMOS spectroscopy",
+                "Subaru/Suprime-Cam imaging"
+            ],
+            "instrument": [
+                "VLT/VIMOS",
+                "Subaru/Suprime-Cam"
+            ],
+            "object": [
+                "cluster A1689"
+            ]
+        },
+        {
+            "property": [
+                "radial velocity distribution",
+                "amplitude",
+                "velocity",
+                "radius"
+            ],
+            "object": [
+                "cluster members",
+                "velocity caustics"
+            ]
+        },
+        {
+            "model": [
+                "Jeans equation"
+            ],
+            "method": [
+                "model-independent method"
+            ],
+            "object": [
+                "velocity anisotropy",
+                "galaxy",
+                "velocity dispersion",
+                "galaxy counts",
+                "cluster mass profile"
+            ],
+            "property": [
+                "3D"
+            ]
+        },
+        {
+            "property": [
+                "velocity anisotropy"
+            ]
+        },
+        {
+            "method": [
+                "solution of the Jeans equation",
+                "method"
+            ],
+            "object": [
+                "galaxy",
+                "mass distribution",
+                "caustic amplitude",
+                "escape velocity"
+            ]
+        },
+        {
+            "model": [
+                "dynamical methods"
+            ],
+            "object": [
+                "cluster virial mass",
+                "lensing",
+                "X-ray analysis"
+            ]
+        },
+        {
+            "property": [
+                "NFW concentration parameter"
+            ]
+        },
+        {
+            "object": [
+                "virial radius",
+                "caustics",
+                "cluster members",
+                "A1689"
+            ],
+            "property": [
+                "velocity"
+            ]
+        },
+        {
+            "object": [
+                "neutronization",
+                "core-collapse supernovae"
+            ]
+        },
+        {
+            "property": [
+                "temperature dependence",
+                "nuclear symmetry energy"
+            ],
+            "object": [
+                "stellar core"
+            ]
+        },
+        {
+            "model": [
+                "one-zone simulation"
+            ],
+            "task": [
+                "follow the collapse"
+            ]
+        },
+        {
+            "property": [
+                "strength"
+            ]
+        },
+        {
+            "property": [
+                "temperature dependence",
+                "symmetry energy",
+                "dissociation energy",
+                "shock wave energetics"
+            ]
+        },
+        {
+            "model": [
+                "one-zone model"
+            ]
+        },
+        {
+            "object": [
+                "Hadrons",
+                "Kerr-Newman Black Holes"
+            ]
+        },
+        {
+            "model": [
+                "Einstein field equations"
+            ]
+        },
+        {
+            "model": [
+                "hypothesis"
+            ],
+            "object": [
+                "proton",
+                "Kerr black hole"
+            ],
+            "property": [
+                "mass",
+                "angular momentum"
+            ]
+        },
+        {
+            "model": [
+                "Kerr-Newman solution of the Einstein-Maxwell equations"
+            ],
+            "object": [
+                "proton"
+            ]
+        },
+        {
+            "object": [
+                "alpha particle"
+            ],
+            "property": [
+                "radius"
+            ]
+        },
+        {
+            "model": [
+                "Kerr solution of General Relativity"
+            ],
+            "object": [
+                "subatomic particle mass spectrum"
+            ]
+        },
+        {
+            "property": [
+                "particle masses",
+                "Kerr solution",
+                "Planck mass"
+            ],
+            "object": [
+                "particle"
+            ]
+        },
+        {
+            "property": [
+                "accuracy"
+            ]
+        },
+        {
+            "property": [
+                "fine structure constant"
+            ]
+        },
+        {
+            "model": [
+                "Large-Scale Magnetic-Field Generation"
+            ]
+        },
+        {
+            "model": [
+                "rigorous theory"
+            ],
+            "method": [
+                "random non-helically forced motions"
+            ],
+            "property": [
+                "large-scale",
+                "low Rm",
+                "weak shear"
+            ],
+            "object": [
+                "magnetic field",
+                "conducting fluid"
+            ]
+        },
+        {
+            "object": [
+                "dynamo",
+                "electromotive force"
+            ]
+        },
+        {
+            "model": [
+                "quasilinear calculation",
+                "shear dynamo"
+            ]
+        },
+        {
+            "method": [
+                "analytically"
+            ]
+        },
+        {
+            "model": [
+                "shear dynamo action"
+            ]
+        },
+        {
+            "object": [
+                "Bullet Cluster",
+                "Sunyaev-Zel'dovich Effect"
+            ]
+        },
+        {
+            "instrument": [
+                "APEX-SZ"
+            ],
+            "object": [
+                "Sunyaev-Zel'dovich effect (SZE)",
+                "Bullet cluster"
+            ]
+        },
+        {
+            "object": [
+                "SZE",
+                "galaxy cluster"
+            ]
+        },
+        {
+            "property": [
+                "significance"
+            ]
+        },
+        {
+            "object": [
+                "SZE map",
+                "X-ray maps",
+                "SZE emission",
+                "radio sources",
+                "IR sources"
+            ]
+        },
+        {
+            "model": [
+                "isothermal elliptical beta model"
+            ]
+        },
+        {
+            "property": [
+                "power-law index",
+                "beta",
+                "core radius",
+                "axial ratio",
+                "central temperature decrement",
+                "flux calibration uncertainty"
+            ]
+        },
+        {
+            "dataset": [
+                "APEX-SZ map",
+                "Chandra X-ray observations"
+            ],
+            "instrument": [
+                "APEX-SZ",
+                "Chandra"
+            ],
+            "property": [
+                "mass-weighted temperature",
+                "X-ray spectroscopic temperatures"
+            ]
+        },
+        {
+            "model": [
+                "isothermal cluster gas distribution"
+            ],
+            "object": [
+                "gas",
+                "cluster",
+                "X-ray"
+            ]
+        },
+        {
+            "instrument": [
+                "APEX-SZ experiment"
+            ]
+        },
+        {
+            "field": [
+                "Curved Space-time"
+            ]
+        },
+        {
+            "field": [
+                "space-time"
+            ]
+        },
+        {
+            "property": [
+                "time"
+            ]
+        },
+        {},
+        {},
+        {
+            "method": [
+                "constructing method"
+            ]
+        },
+        {
+            "model": [
+                "classical approximation of spinor equation",
+                "Einstein's equation"
+            ],
+            "method": [
+                "classical approximation",
+                "weak field approximation",
+                "low speed approximation"
+            ]
+        },
+        {
+            "model": [
+                "coordinate system"
+            ]
+        },
+        {},
+        {
+            "object": [
+                "Tully-Fisher Relation"
+            ]
+        },
+        {
+            "dataset": [
+                "2 Micron All-Sky Survey (2MASS)",
+                "2MASS Redshift Survey (2MRS)"
+            ],
+            "instrument": [
+                "2 Micron All-Sky Survey (2MASS)"
+            ],
+            "object": [
+                "Tully-Fisher distances",
+                "spirals"
+            ]
+        },
+        {
+            "method": [
+                "TF relation"
+            ]
+        },
+        {
+            "model": [
+                "TF template"
+            ]
+        },
+        {
+            "property": [
+                "slope",
+                "wavelength"
+            ],
+            "object": [
+                "TF relation",
+                "K-band",
+                "J",
+                "H-bands"
+            ]
+        },
+        {
+            "object": [
+                "galaxy",
+                "spirals"
+            ],
+            "property": [
+                "morphology"
+            ]
+        },
+        {
+            "field": [
+                "Sc galaxies"
+            ]
+        },
+        {
+            "property": [
+                "width",
+                "intrinsic scatter"
+            ]
+        },
+        {
+            "object": [
+                "S1 Shell",
+                "Interstellar Magnetic Field",
+                "Gas",
+                "Heliosphere"
+            ]
+        },
+        {
+            "object": [
+                "Loop I magnetic superbubble",
+                "Sun"
+            ]
+        },
+        {
+            "model": [
+                "magnetic shells"
+            ],
+            "object": [
+                "Loop I",
+                "Sun",
+                "'S1' shell"
+            ]
+        },
+        {
+            "object": [
+                "Sun",
+                "interstellar magnetic field",
+                "interstellar material"
+            ]
+        },
+        {
+            "object": [
+                "subshells",
+                "interstellar magnetic field (ISMF)",
+                "interstellar FeII",
+                "CaII",
+                "Sun"
+            ],
+            "property": [
+                "properties"
+            ]
+        },
+        {
+            "model": [
+                "Loop I"
+            ],
+            "object": [
+                "ISM"
+            ]
+        },
+        {
+            "object": [
+                "CaII",
+                "interstellar radiation field",
+                "star",
+                "magnetic subshells"
+            ],
+            "property": [
+                "column densities",
+                "strength"
+            ]
+        },
+        {
+            "object": [
+                "dipole moment",
+                "cosmic microwave background",
+                "heliosphere",
+                "local ISMF"
+            ]
+        },
+        {
+            "object": [
+                "supermassive black holes",
+                "pseudo-bulges",
+                "classical bulges",
+                "elliptical galaxies"
+            ]
+        },
+        {
+            "dataset": [
+                "Sloan Digital Sky Survey"
+            ],
+            "object": [
+                "galaxies",
+                "elliptical galaxies",
+                "classical bulges",
+                "pseudo-bulges",
+                "black holes"
+            ],
+            "property": [
+                "mass",
+                "stellar masses",
+                "central stellar velocity dispersion"
+            ]
+        },
+        {
+            "object": [
+                "elliptical galaxy",
+                "bulge",
+                "black hole",
+                "local universe",
+                "classical bulges",
+                "pseudo-bulges"
+            ],
+            "property": [
+                "mass"
+            ]
+        },
+        {
+            "object": [
+                "ellipticals",
+                "classical bulges",
+                "pseudo-bulges",
+                "stellar masses",
+                "sigma",
+                "barred galaxies"
+            ]
+        },
+        {
+            "object": [
+                "black hole",
+                "M_BH-M_Bulge",
+                "M_BH-sigma relations"
+            ]
+        },
+        {
+            "property": [
+                "total mass density"
+            ]
+        },
+        {
+            "property": [
+                "Characteristic Scales",
+                "Initial Density",
+                "Velocity Fields"
+            ]
+        },
+        {
+            "property": [
+                "density contrast",
+                "peculiar velocity",
+                "differential length scales",
+                "integral length scales"
+            ],
+            "object": [
+                "field"
+            ]
+        },
+        {
+            "object": [
+                "halos",
+                "galaxies",
+                "galaxy distribution"
+            ]
+        },
+        {
+            "object": [
+                "scales",
+                "fields"
+            ]
+        },
+        {
+            "object": [
+                "Gravitino Dark Matter",
+                "Flavour Structure",
+                "R-violating Operators"
+            ]
+        },
+        {
+            "model": [
+                "R-violating supersymmetry"
+            ],
+            "object": [
+                "gravitino dark matter",
+                "gravitino"
+            ]
+        },
+        {
+            "property": [
+                "dominant decay modes",
+                "R-violating hierarchies"
+            ]
+        },
+        {
+            "property": [
+                "relative magnitude",
+                "phase space suppression"
+            ]
+        },
+        {
+            "object": [
+                "gravitino",
+                "Next-to-Lightest Supersymmetric Particle (NLSP)",
+                "superparticle"
+            ]
+        },
+        {
+            "object": [
+                "Dark Matter Caustics",
+                "Galaxy Clusters"
+            ]
+        },
+        {
+            "object": [
+                "gravitational lensing observations",
+                "Jee"
+            ]
+        },
+        {
+            "object": [
+                "caustic ring",
+                "dark matter",
+                "galaxy cluster"
+            ]
+        },
+        {
+            "model": [
+                "gravitational potential well"
+            ]
+        },
+        {
+            "object": [
+                "caustic rings",
+                "dark matter",
+                "Milky Way",
+                "spiral galaxies"
+            ]
+        },
+        {
+            "object": [
+                "galaxy clusters",
+                "caustic rings"
+            ]
+        },
+        {
+            "object": [
+                "caustic ring",
+                "cluster"
+            ],
+            "property": [
+                "column density profile"
+            ]
+        },
+        {
+            "method": [
+                "radiative transfer"
+            ]
+        },
+        {
+            "model": [
+                "Boltzmann equation"
+            ]
+        },
+        {
+            "model": [
+                "tensor valued distribution function",
+                "gauge-invariant distribution function"
+            ],
+            "object": [
+                "polarized radiation",
+                "distribution function"
+            ]
+        },
+        {
+            "method": [
+                "Liouville operator",
+                "collision term"
+            ]
+        },
+        {
+            "modality": [
+                "multipoles",
+                "normal modes"
+            ]
+        },
+        {
+            "model": [
+                "cosmic microwave background"
+            ],
+            "object": [
+                "perturbations",
+                "non-Gaussianity",
+                "evolution"
+            ]
+        },
+        {
+            "object": [
+                "XO-5b",
+                "Jupiter"
+            ]
+        },
+        {
+            "dataset": [
+                "HATNet observations"
+            ],
+            "object": [
+                "XO-5b",
+                "planetary nature"
+            ]
+        },
+        {
+            "property": [
+                "spectral bisector variations",
+                "orbital phase"
+            ]
+        },
+        {
+            "dataset": [
+                "spectroscopic measurements"
+            ]
+        },
+        {
+            "object": [
+                "planet signals",
+                "data"
+            ]
+        },
+        {
+            "object": [
+                "S activity index",
+                "stellar activity"
+            ]
+        },
+        {
+            "object": [
+                "planet parameters"
+            ]
+        },
+        {},
+        {
+            "object": [
+                "XO-5b",
+                "star"
+            ],
+            "property": [
+                "mass",
+                "radius",
+                "metallicity"
+            ]
+        },
+        {
+            "property": [
+                "planetary mass",
+                "radius",
+                "mean density"
+            ],
+            "object": [
+                "planet"
+            ]
+        },
+        {
+            "property": [
+                "P",
+                "E",
+                "transit duration"
+            ]
+        },
+        {
+            "property": [
+                "transit timing variations"
+            ]
+        },
+        {
+            "object": [
+                "XO-5b",
+                "exoplanets"
+            ],
+            "property": [
+                "Safronov number",
+                "surface gravity"
+            ]
+        },
+        {
+            "object": [
+                "Electron Temperatures",
+                "W51 Complex"
+            ]
+        },
+        {
+            "object": [
+                "W51",
+                "Sagitarius arm",
+                "Sun "
+            ],
+            "property": [
+                "distance",
+                "extension"
+            ]
+        },
+        {
+            "object": [
+                "W51A",
+                "W51B",
+                "W51C",
+                "HII regions",
+                "supernova remnant"
+            ]
+        },
+        {
+            "dataset": [
+                "continuum radio observations"
+            ],
+            "instrument": [
+                "GMRT"
+            ],
+            "modality": [
+                "radio"
+            ],
+            "object": [
+                "HII regions",
+                "W51 complex"
+            ]
+        },
+        {
+            "object": [
+                "W51",
+                "thermal subcomponents"
+            ],
+            "property": [
+                "electron temperatures",
+                "emission measures"
+            ],
+            "modality": [
+                "spectra"
+            ]
+        },
+        {
+            "instrument": [
+                "continuum spectra",
+                "radio recombination line (RRL)"
+            ],
+            "modality": [
+                "spectra"
+            ],
+            "property": [
+                "electron temperatures",
+                "filling factor"
+            ]
+        },
+        {
+            "model": [
+                "multicomponent model"
+            ]
+        },
+        {
+            "object": [
+                "A1689",
+                "Mass Profile",
+                "Concentration Parameter",
+                "Velocity Anisotropy"
+            ]
+        },
+        {
+            "dataset": [
+                "VLT/VIMOS spectroscopy",
+                "Subaru/Suprime-Cam imaging"
+            ],
+            "instrument": [
+                "VLT/VIMOS",
+                "Subaru/Suprime-Cam"
+            ],
+            "object": [
+                "cluster A1689"
+            ]
+        },
+        {
+            "property": [
+                "radial velocity distribution",
+                "amplitude",
+                "velocity",
+                "radius"
+            ],
+            "object": [
+                "cluster members",
+                "velocity caustics"
+            ]
+        },
+        {
+            "model": [
+                "Jeans equation"
+            ],
+            "method": [
+                "model-independent method"
+            ],
+            "object": [
+                "velocity anisotropy",
+                "galaxy",
+                "velocity dispersion",
+                "galaxy counts",
+                "cluster mass profile"
+            ],
+            "property": [
+                "3D"
+            ]
+        },
+        {
+            "property": [
+                "velocity anisotropy"
+            ]
+        },
+        {
+            "method": [
+                "solution of the Jeans equation",
+                "method"
+            ],
+            "object": [
+                "galaxy",
+                "mass distribution",
+                "caustic amplitude",
+                "escape velocity"
+            ]
+        },
+        {
+            "model": [
+                "dynamical methods"
+            ],
+            "object": [
+                "cluster virial mass",
+                "lensing",
+                "X-ray analysis"
+            ]
+        },
+        {
+            "property": [
+                "NFW concentration parameter"
+            ]
+        },
+        {
+            "object": [
+                "virial radius",
+                "caustics",
+                "cluster members",
+                "A1689"
+            ],
+            "property": [
+                "velocity"
+            ]
+        },
+        {
+            "object": [
+                "neutronization",
+                "core-collapse supernovae"
+            ]
+        },
+        {
+            "property": [
+                "temperature dependence",
+                "nuclear symmetry energy"
+            ],
+            "object": [
+                "stellar core"
+            ]
+        },
+        {
+            "model": [
+                "one-zone simulation"
+            ],
+            "task": [
+                "follow the collapse"
+            ]
+        },
+        {
+            "property": [
+                "strength"
+            ]
+        },
+        {
+            "property": [
+                "temperature dependence",
+                "symmetry energy",
+                "dissociation energy",
+                "shock wave energetics"
+            ]
+        },
+        {
+            "model": [
+                "one-zone model"
+            ]
+        },
+        {
+            "object": [
+                "Hadrons",
+                "Kerr-Newman Black Holes"
+            ]
+        },
+        {
+            "model": [
+                "Einstein field equations"
+            ]
+        },
+        {
+            "model": [
+                "hypothesis"
+            ],
+            "object": [
+                "proton",
+                "Kerr black hole"
+            ],
+            "property": [
+                "mass",
+                "angular momentum"
+            ]
+        },
+        {
+            "model": [
+                "Kerr-Newman solution of the Einstein-Maxwell equations"
+            ],
+            "object": [
+                "proton"
+            ]
+        },
+        {
+            "object": [
+                "alpha particle"
+            ],
+            "property": [
+                "radius"
+            ]
+        },
+        {
+            "model": [
+                "Kerr solution of General Relativity"
+            ],
+            "object": [
+                "subatomic particle mass spectrum"
+            ]
+        },
+        {
+            "property": [
+                "particle masses",
+                "Kerr solution",
+                "Planck mass"
+            ],
+            "object": [
+                "particle"
+            ]
+        },
+        {
+            "property": [
+                "accuracy"
+            ]
+        },
+        {
+            "property": [
+                "fine structure constant"
+            ]
+        },
+        {
+            "model": [
+                "Large-Scale Magnetic-Field Generation"
+            ]
+        },
+        {
+            "model": [
+                "rigorous theory"
+            ],
+            "method": [
+                "random non-helically forced motions"
+            ],
+            "property": [
+                "large-scale",
+                "low Rm",
+                "weak shear"
+            ],
+            "object": [
+                "magnetic field",
+                "conducting fluid"
+            ]
+        },
+        {
+            "object": [
+                "dynamo",
+                "electromotive force"
+            ]
+        },
+        {
+            "model": [
+                "quasilinear calculation",
+                "shear dynamo"
+            ]
+        },
+        {
+            "method": [
+                "analytically"
+            ]
+        },
+        {
+            "model": [
+                "shear dynamo action"
+            ]
+        },
+        {
+            "object": [
+                "Bullet Cluster",
+                "Sunyaev-Zel'dovich Effect"
+            ]
+        },
+        {
+            "instrument": [
+                "APEX-SZ"
+            ],
+            "object": [
+                "Sunyaev-Zel'dovich effect (SZE)",
+                "Bullet cluster"
+            ]
+        },
+        {
+            "object": [
+                "SZE",
+                "galaxy cluster"
+            ]
+        },
+        {
+            "property": [
+                "significance"
+            ]
+        },
+        {
+            "object": [
+                "SZE map",
+                "X-ray maps",
+                "SZE emission",
+                "radio sources",
+                "IR sources"
+            ]
+        },
+        {
+            "model": [
+                "isothermal elliptical beta model"
+            ]
+        },
+        {
+            "property": [
+                "power-law index",
+                "beta",
+                "core radius",
+                "axial ratio",
+                "central temperature decrement",
+                "flux calibration uncertainty"
+            ]
+        },
+        {
+            "dataset": [
+                "APEX-SZ map",
+                "Chandra X-ray observations"
+            ],
+            "instrument": [
+                "APEX-SZ",
+                "Chandra"
+            ],
+            "property": [
+                "mass-weighted temperature",
+                "X-ray spectroscopic temperatures"
+            ]
+        },
+        {
+            "model": [
+                "isothermal cluster gas distribution"
+            ],
+            "object": [
+                "gas",
+                "cluster",
+                "X-ray"
+            ]
+        },
+        {
+            "instrument": [
+                "APEX-SZ experiment"
+            ]
+        },
+        {
+            "field": [
+                "Curved Space-time"
+            ]
+        },
+        {
+            "field": [
+                "space-time"
+            ]
+        },
+        {
+            "property": [
+                "time"
+            ]
+        },
+        {},
+        {},
+        {
+            "method": [
+                "constructing method"
+            ]
+        },
+        {
+            "model": [
+                "classical approximation of spinor equation",
+                "Einstein's equation"
+            ],
+            "method": [
+                "classical approximation",
+                "weak field approximation",
+                "low speed approximation"
+            ]
+        },
+        {
+            "model": [
+                "coordinate system"
+            ]
+        },
+        {},
+        {
+            "object": [
+                "Tully-Fisher Relation"
+            ]
+        },
+        {
+            "dataset": [
+                "2 Micron All-Sky Survey (2MASS)",
+                "2MASS Redshift Survey (2MRS)"
+            ],
+            "instrument": [
+                "2 Micron All-Sky Survey (2MASS)"
+            ],
+            "object": [
+                "Tully-Fisher distances",
+                "spirals"
+            ]
+        },
+        {
+            "method": [
+                "TF relation"
+            ]
+        },
+        {
+            "model": [
+                "TF template"
+            ]
+        },
+        {
+            "property": [
+                "slope",
+                "wavelength"
+            ],
+            "object": [
+                "TF relation",
+                "K-band",
+                "J",
+                "H-bands"
+            ]
+        },
+        {
+            "object": [
+                "galaxy",
+                "spirals"
+            ],
+            "property": [
+                "morphology"
+            ]
+        },
+        {
+            "field": [
+                "Sc galaxies"
+            ]
+        },
+        {
+            "property": [
+                "width",
+                "intrinsic scatter"
+            ]
+        },
+        {
+            "object": [
+                "S1 Shell",
+                "Interstellar Magnetic Field",
+                "Gas",
+                "Heliosphere"
+            ]
+        },
+        {
+            "object": [
+                "Loop I magnetic superbubble",
+                "Sun"
+            ]
+        },
+        {
+            "model": [
+                "magnetic shells"
+            ],
+            "object": [
+                "Loop I",
+                "Sun",
+                "'S1' shell"
+            ]
+        },
+        {
+            "object": [
+                "Sun",
+                "interstellar magnetic field",
+                "interstellar material"
+            ]
+        },
+        {
+            "object": [
+                "subshells",
+                "interstellar magnetic field (ISMF)",
+                "interstellar FeII",
+                "CaII",
+                "Sun"
+            ],
+            "property": [
+                "properties"
+            ]
+        },
+        {
+            "model": [
+                "Loop I"
+            ],
+            "object": [
+                "ISM"
+            ]
+        },
+        {
+            "object": [
+                "CaII",
+                "interstellar radiation field",
+                "star",
+                "magnetic subshells"
+            ],
+            "property": [
+                "column densities",
+                "strength"
+            ]
+        },
+        {
+            "object": [
+                "dipole moment",
+                "cosmic microwave background",
+                "heliosphere",
+                "local ISMF"
+            ]
+        },
+        {
+            "object": [
+                "supermassive black holes",
+                "pseudo-bulges",
+                "classical bulges",
+                "elliptical galaxies"
+            ]
+        },
+        {
+            "dataset": [
+                "Sloan Digital Sky Survey"
+            ],
+            "object": [
+                "galaxies",
+                "elliptical galaxies",
+                "classical bulges",
+                "pseudo-bulges",
+                "black holes"
+            ],
+            "property": [
+                "mass",
+                "stellar masses",
+                "central stellar velocity dispersion"
+            ]
+        },
+        {
+            "object": [
+                "elliptical galaxy",
+                "bulge",
+                "black hole",
+                "local universe",
+                "classical bulges",
+                "pseudo-bulges"
+            ],
+            "property": [
+                "mass"
+            ]
+        },
+        {
+            "object": [
+                "ellipticals",
+                "classical bulges",
+                "pseudo-bulges",
+                "stellar masses",
+                "sigma",
+                "barred galaxies"
+            ]
+        },
+        {
+            "object": [
+                "black hole",
+                "M_BH-M_Bulge",
+                "M_BH-sigma relations"
+            ]
+        },
+        {
+            "property": [
+                "total mass density"
+            ]
+        },
+        {
+            "property": [
+                "Characteristic Scales",
+                "Initial Density",
+                "Velocity Fields"
+            ]
+        },
+        {
+            "property": [
+                "density contrast",
+                "peculiar velocity",
+                "differential length scales",
+                "integral length scales"
+            ],
+            "object": [
+                "field"
+            ]
+        },
+        {
+            "object": [
+                "halos",
+                "galaxies",
+                "galaxy distribution"
+            ]
+        },
+        {
+            "object": [
+                "scales",
+                "fields"
+            ]
+        },
+        {
+            "object": [
+                "Gravitino Dark Matter",
+                "Flavour Structure",
+                "R-violating Operators"
+            ]
+        },
+        {
+            "model": [
+                "R-violating supersymmetry"
+            ],
+            "object": [
+                "gravitino dark matter",
+                "gravitino"
+            ]
+        },
+        {
+            "property": [
+                "dominant decay modes",
+                "R-violating hierarchies"
+            ]
+        },
+        {
+            "property": [
+                "relative magnitude",
+                "phase space suppression"
+            ]
+        },
+        {
+            "object": [
+                "gravitino",
+                "Next-to-Lightest Supersymmetric Particle (NLSP)",
+                "superparticle"
+            ]
+        },
+        {
+            "object": [
+                "Dark Matter Caustics",
+                "Galaxy Clusters"
+            ]
+        },
+        {
+            "object": [
+                "gravitational lensing observations",
+                "Jee"
+            ]
+        },
+        {
+            "object": [
+                "caustic ring",
+                "dark matter",
+                "galaxy cluster"
+            ]
+        },
+        {
+            "model": [
+                "gravitational potential well"
+            ]
+        },
+        {
+            "object": [
+                "caustic rings",
+                "dark matter",
+                "Milky Way",
+                "spiral galaxies"
+            ]
+        },
+        {
+            "object": [
+                "galaxy clusters",
+                "caustic rings"
+            ]
+        },
+        {
+            "object": [
+                "caustic ring",
+                "cluster"
+            ],
+            "property": [
+                "column density profile"
+            ]
+        },
+        {
+            "field": [
+                "dark matter"
+            ]
+        },
+        {
+            "task": [
+                "estimating the mass"
+            ]
+        },
+        {
+            "model": [
+                "4-component model"
+            ],
+            "object": [
+                "elliptical galaxies",
+                "stars",
+                "dark matter (DM)",
+                "hot gas",
+                "central black hole"
+            ]
+        },
+        {
+            "object": [
+                "DM",
+                "stars",
+                "central black hole"
+            ]
+        },
+        {
+            "model": [
+                "DM density profile"
+            ]
+        },
+        {
+            "property": [
+                "gas fraction",
+                "baryon fraction"
+            ],
+            "object": [
+                "radius",
+                "effective radii (R_e)"
+            ]
+        },
+        {
+            "property": [
+                "line-of-sight velocity dispersion",
+                "velocity anisotropy",
+                "total mass"
+            ],
+            "object": [
+                "velocity"
+            ]
+        },
+        {
+            "model": [
+                "LCDM simulations"
+            ],
+            "object": [
+                "DM distributions",
+                "stellar VD profiles",
+                "VDs"
+            ]
+        },
+        {
+            "object": [
+                "planetary nebulae",
+                "baryon fraction"
+            ]
+        },
+        {
+            "model": [
+                "NFW DM model",
+                "Navarro et al"
+            ]
+        },
+        {},
+        {
+            "model": [
+                "NFW model"
+            ]
+        },
+        {
+            "model": [
+                "LCDM",
+                "NFW"
+            ],
+            "property": [
+                "radial anisotropy",
+                "M/L"
+            ]
+        },
+        {
+            "model": [
+                "DM model"
+            ]
+        },
+        {
+            "model": [
+                "anisotropic models",
+                "Osipkov-Merritt model"
+            ]
+        },
+        {
+            "object": [
+                "hyperons",
+                "neutron stars"
+            ]
+        },
+        {
+            "model": [
+                "equation of state"
+            ],
+            "object": [
+                "hyperon-nucleon potential",
+                "nucleonic matter"
+            ]
+        },
+        {
+            "method": [
+                "renormalization group framework"
+            ],
+            "object": [
+                "hyperon-nucleon potential"
+            ]
+        },
+        {
+            "model": [
+                "Hartree-Fock approximation"
+            ]
+        },
+        {
+            "object": [
+                "hyperons",
+                "neutron stars"
+            ]
+        },
+        {
+            "object": [
+                "neutron stars"
+            ]
+        }
+    ]
+}

data/results/new_random_json_QdeCg2hM_20240801_034716/logs/log.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:224923a572be02a4cd66af132c9ce4360615664a787b5fb1922fe927b3072f49
+size 177287528

data/results/new_random_json_QdeCg2hM_20240801_034716/metrics.json

@@ -0,0 +1,7 @@
+{
+    "metrics": {
+        "avg_pred_response_time_per_sentence": 6.1593,
+        "total_time": 227749.3056
+    },
+    "prompt_file": "prompts.txt"
+}