Quantization made by Richard Erkhov.
[Github](https://github.com/RichardErkhov)
[Discord](https://discord.gg/pvy7H8DZMG)
[Request more models](https://github.com/RichardErkhov/quant_request)
GNER-LLaMA-7B - GGUF
- Model creator: https://huggingface.co/dyyyyyyyy/
- Original model: https://huggingface.co/dyyyyyyyy/GNER-LLaMA-7B/
| Name | Quant method | Size |
| ---- | ---- | ---- |
| [GNER-LLaMA-7B.Q2_K.gguf](https://huggingface.co/RichardErkhov/dyyyyyyyy_-_GNER-LLaMA-7B-gguf/blob/main/GNER-LLaMA-7B.Q2_K.gguf) | Q2_K | 2.36GB |
| [GNER-LLaMA-7B.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/dyyyyyyyy_-_GNER-LLaMA-7B-gguf/blob/main/GNER-LLaMA-7B.IQ3_XS.gguf) | IQ3_XS | 2.6GB |
| [GNER-LLaMA-7B.IQ3_S.gguf](https://huggingface.co/RichardErkhov/dyyyyyyyy_-_GNER-LLaMA-7B-gguf/blob/main/GNER-LLaMA-7B.IQ3_S.gguf) | IQ3_S | 2.75GB |
| [GNER-LLaMA-7B.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/dyyyyyyyy_-_GNER-LLaMA-7B-gguf/blob/main/GNER-LLaMA-7B.Q3_K_S.gguf) | Q3_K_S | 2.75GB |
| [GNER-LLaMA-7B.IQ3_M.gguf](https://huggingface.co/RichardErkhov/dyyyyyyyy_-_GNER-LLaMA-7B-gguf/blob/main/GNER-LLaMA-7B.IQ3_M.gguf) | IQ3_M | 2.9GB |
| [GNER-LLaMA-7B.Q3_K.gguf](https://huggingface.co/RichardErkhov/dyyyyyyyy_-_GNER-LLaMA-7B-gguf/blob/main/GNER-LLaMA-7B.Q3_K.gguf) | Q3_K | 3.07GB |
| [GNER-LLaMA-7B.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/dyyyyyyyy_-_GNER-LLaMA-7B-gguf/blob/main/GNER-LLaMA-7B.Q3_K_M.gguf) | Q3_K_M | 3.07GB |
| [GNER-LLaMA-7B.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/dyyyyyyyy_-_GNER-LLaMA-7B-gguf/blob/main/GNER-LLaMA-7B.Q3_K_L.gguf) | Q3_K_L | 3.35GB |
| [GNER-LLaMA-7B.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/dyyyyyyyy_-_GNER-LLaMA-7B-gguf/blob/main/GNER-LLaMA-7B.IQ4_XS.gguf) | IQ4_XS | 3.4GB |
| [GNER-LLaMA-7B.Q4_0.gguf](https://huggingface.co/RichardErkhov/dyyyyyyyy_-_GNER-LLaMA-7B-gguf/blob/main/GNER-LLaMA-7B.Q4_0.gguf) | Q4_0 | 3.56GB |
| [GNER-LLaMA-7B.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/dyyyyyyyy_-_GNER-LLaMA-7B-gguf/blob/main/GNER-LLaMA-7B.IQ4_NL.gguf) | IQ4_NL | 3.58GB |
| [GNER-LLaMA-7B.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/dyyyyyyyy_-_GNER-LLaMA-7B-gguf/blob/main/GNER-LLaMA-7B.Q4_K_S.gguf) | Q4_K_S | 3.59GB |
| [GNER-LLaMA-7B.Q4_K.gguf](https://huggingface.co/RichardErkhov/dyyyyyyyy_-_GNER-LLaMA-7B-gguf/blob/main/GNER-LLaMA-7B.Q4_K.gguf) | Q4_K | 3.8GB |
| [GNER-LLaMA-7B.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/dyyyyyyyy_-_GNER-LLaMA-7B-gguf/blob/main/GNER-LLaMA-7B.Q4_K_M.gguf) | Q4_K_M | 3.8GB |
| [GNER-LLaMA-7B.Q4_1.gguf](https://huggingface.co/RichardErkhov/dyyyyyyyy_-_GNER-LLaMA-7B-gguf/blob/main/GNER-LLaMA-7B.Q4_1.gguf) | Q4_1 | 3.95GB |
| [GNER-LLaMA-7B.Q5_0.gguf](https://huggingface.co/RichardErkhov/dyyyyyyyy_-_GNER-LLaMA-7B-gguf/blob/main/GNER-LLaMA-7B.Q5_0.gguf) | Q5_0 | 4.33GB |
| [GNER-LLaMA-7B.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/dyyyyyyyy_-_GNER-LLaMA-7B-gguf/blob/main/GNER-LLaMA-7B.Q5_K_S.gguf) | Q5_K_S | 4.33GB |
| [GNER-LLaMA-7B.Q5_K.gguf](https://huggingface.co/RichardErkhov/dyyyyyyyy_-_GNER-LLaMA-7B-gguf/blob/main/GNER-LLaMA-7B.Q5_K.gguf) | Q5_K | 4.45GB |
| [GNER-LLaMA-7B.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/dyyyyyyyy_-_GNER-LLaMA-7B-gguf/blob/main/GNER-LLaMA-7B.Q5_K_M.gguf) | Q5_K_M | 4.45GB |
| [GNER-LLaMA-7B.Q5_1.gguf](https://huggingface.co/RichardErkhov/dyyyyyyyy_-_GNER-LLaMA-7B-gguf/blob/main/GNER-LLaMA-7B.Q5_1.gguf) | Q5_1 | 4.72GB |
| [GNER-LLaMA-7B.Q6_K.gguf](https://huggingface.co/RichardErkhov/dyyyyyyyy_-_GNER-LLaMA-7B-gguf/blob/main/GNER-LLaMA-7B.Q6_K.gguf) | Q6_K | 5.15GB |
| [GNER-LLaMA-7B.Q8_0.gguf](https://huggingface.co/RichardErkhov/dyyyyyyyy_-_GNER-LLaMA-7B-gguf/blob/main/GNER-LLaMA-7B.Q8_0.gguf) | Q8_0 | 6.67GB |
Original model description:
---
license: apache-2.0
datasets:
- Universal-NER/Pile-NER-type
language:
- en
metrics:
- f1
library_name: transformers
pipeline_tag: text-generation
---
Rethinking Negative Instances for Generative Named Entity Recognition
# Model Card for GNER-LLaMA-7B
We introduce GNER, a **G**enerative **N**amed **E**ntity **R**ecognition framework, which demonstrates enhanced zero-shot capabilities across unseen entity domains. Experiments on two representative generative models, i.e., LLaMA and Flan-T5, show that the integration of negative instances into the training process yields substantial performance enhancements. The resulting models, GNER-LLaMA and GNER-T5, outperform state-of-the-art (SoTA) approaches by a large margin, achieving improvements of 8 and 11 points in $F_1$ score, respectively. Code and models are publicly available.
* ๐ป Code: [https://github.com/yyDing1/GNER/](https://github.com/yyDing1/GNER/)
* ๐ Paper: [Rethinking Negative Instances for Generative Named Entity Recognition](https://arxiv.org/abs/2402.16602)
* ๐พ Models in the ๐ค HuggingFace Hub: [GNER-Models](https://huggingface.co/collections/dyyyyyyyy/gner-65dda2cb96c6e35c814dea56)
* ๐งช Reproduction Materials: [Reproduction Materials](https://drive.google.com/drive/folders/1m2FqDgItEbSoeUVo-i18AwMvBcNkZD46?usp=drive_link)
* ๐จ Example Jupyter Notebooks: [GNER Notebook](https://github.com/yyDing1/GNER/blob/main/notebook.ipynb)
## PreTrained Models
We release five GNER models based on LLaMA (7B) and Flan-T5 (base, large, xl and xxl).
| Model | # Params | Zero-shot Average $F_1$ | Supervised Average $F_1$ | ๐ค HuggingFace
Download Link |
| ------------- | -------: | :----------------------: | :-----------------------: | :-------------------------------------------------: |
| GNER-LLaMA | 7B | 66.1 | 86.09 | [link](https://huggingface.co/dyyyyyyyy/GNER-LLaMA-7B) |
| GNER-T5-base | 248M | 59.5 | 83.21 | [link](https://huggingface.co/dyyyyyyyy/GNER-T5-base) |
| GNER-T5-large | 783M | 63.5 | 85.45 | [link](https://huggingface.co/dyyyyyyyy/GNER-T5-large) |
| GNER-T5-xl | 3B | 66.1 | 85.94 | [link](https://huggingface.co/dyyyyyyyy/GNER-T5-xl) |
| GNER-T5-xxl | 11B | 69.1 | 86.15 | [link](https://huggingface.co/dyyyyyyyy/GNER-T5-xxl) |
## Demo usage
You should install the dependencies:
```bash
pip install torch datasets deepspeed accelerate transformers protobuf
```
Please check out [Example Jupyter Notebooks](https://github.com/yyDing1/GNER/blob/main/notebook.ipynb) for guidance on utilizing GNER models.
A simple inference example is as follows:
Below is an example using `GNER-LLaMA`
```python
>>> import torch
>>> from transformers import AutoTokenizer, AutoModelForCausalLM
>>> tokenizer = AutoTokenizer.from_pretrained("dyyyyyyyy/GNER-LLaMA-7B")
>>> model = AutoModelForCausalLM.from_pretrained("dyyyyyyyy/GNER-LLaMA-7B", torch_dtype=torch.bfloat16).cuda()
>>> model = model.eval()
>>> instruction_template = "Please analyze the sentence provided, identifying the type of entity for each word on a token-by-token basis.\nOutput format is: word_1(label_1), word_2(label_2), ...\nWe'll use the BIO-format to label the entities, where:\n1. B- (Begin) indicates the start of a named entity.\n2. I- (Inside) is used for words within a named entity but are not the first word.\n3. O (Outside) denotes words that are not part of a named entity.\n"
>>> sentence = "did george clooney make a musical in the 1980s"
>>> entity_labels = ["genre", "rating", "review", "plot", "song", "average ratings", "director", "character", "trailer", "year", "actor", "title"]
>>> instruction = f"{instruction_template}\nUse the specific entity tags: {', '.join(entity_labels)} and O.\nSentence: {sentence}"
>>> instruction = f"[INST] {instruction} [/INST]"
>>> inputs = tokenizer(instruction, return_tensors="pt").to("cuda")
>>> outputs = model.generate(**inputs, max_new_tokens=640)
>>> response = tokenizer.decode(outputs[0], skip_special_tokens=True)
>>> response = response[response.find("[/INST]") + len("[/INST]"):].strip()
>>> print(response)
"did(O) george(B-actor) clooney(I-actor) make(O) a(O) musical(B-genre) in(O) the(O) 1980s(B-year)"
```
## Citation
```bibtex
@misc{ding2024rethinking,
title={Rethinking Negative Instances for Generative Named Entity Recognition},
author={Yuyang Ding and Juntao Li and Pinzheng Wang and Zecheng Tang and Bowen Yan and Min Zhang},
year={2024},
eprint={2402.16602},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```