Datasets:

Modalities:
Text
ArXiv:
Libraries:
Datasets
davidlvxin commited on
Commit
c2a61c1
1 Parent(s): 10e175d

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +41 -39
README.md CHANGED
@@ -22,6 +22,8 @@ We are fully aware of the potentially high costs involved in the model evaluatio
22
 
23
  LongBench includes 13 English tasks, 5 Chinese tasks, and 2 code tasks, with the average length of most tasks ranging from 5k to 15k.
24
 
 
 
25
  # How to use it?
26
 
27
  #### Loading Data
@@ -61,67 +63,67 @@ This repository provides data download for LongBench. If you wish to use this da
61
  # Task statistics
62
 
63
  | Task | Task Type | Eval metric | Avg len |Language | \#Sample |
64
- | --------- | -------------| ------------- |--------- | ------------- |--------- |
65
- | HotpotQA | Multi-doc QA | F1 |9149 |EN |200 |
66
- | 2WikiMultihopQA| Multi-doc QA | F1 |4885 |EN |200 |
67
- | Musique| Multi-doc QA | F1 |7798 |EN |200 |
68
- | DuReader| Multi-doc QA | Rouge-L |15768 |ZH |200 |
69
- | MultiFieldQA-en| Single-doc QA | F1 |4559 |EN |150 |
70
- | MultiFieldQA-zh| Single-doc QA | F1 |6771 |ZH |200 |
71
- | NarrativeQA| Single-doc QA | F1 |18405 |EN |200 |
72
- | Qasper| Single-doc QA | F1 |3619 |EN |200 |
73
- | GovReport| Summarization | Rouge-L |8169 |EN |200 |
74
- | QMSum| Summarization | Rouge-L |10546 |EN |200 |
75
- | VCSUM| Summarization | Rouge-L |15147 |ZH |200 |
76
- | TriviaQA| Few shot | F1 |8015 |EN |200 |
77
- | NQ| Few shot | F1 |8210 |EN |200 |
78
- | TREC| Few shot | Accuracy |5176 |EN |200 |
79
- | LSHT| Few shot | Accuracy |22333 |ZH |200 |
80
- | PassageRetrieval-en| Synthetic | Accuracy |9288 |EN |200 |
81
- | PassageCount| Synthetic | Accuracy |11141 |EN |200 |
82
- | PassageRetrieval-zh | Synthetic | Accuracy |6745 |ZH |200 |
83
- | LCC| Code | Edit Sim |1235 |Python/C#/Java |500 |
84
- | RepoBench-P| Code | Edit Sim |5622 |Python/Java |500 |
85
 
86
  > Note: In order to avoid discrepancies caused by different tokenizers, we use the word count (using Python's split function) to calculate the average length of English datasets and code datasets, and use the character count to calculate the average length of Chinese datasets.
87
 
88
  # Task description
89
 
90
  | Task | Task Description |
91
- | ----------------- | ------------------------------------------------------------ |
92
  | HotpotQA | Answer related questions based on multiple given documents |
93
  | 2WikiMultihopQA | Answer related questions based on multiple given documents |
94
  | Musique | Answer related questions based on multiple given documents |
95
  | DuReader | Answer related Chinese questions based on multiple retrieved documents |
96
- | MultiFieldQA-en | Answer English questions based on a single document, which comes from a relatively diverse field |
97
- | MultiFieldQA-zh | Answer Chinese questions based on a single document, which comes from a relatively diverse field |
98
  | NarrativeQA | Ask questions based on stories or scripts, including understanding of important elements such as characters, plots, themes, etc. |
99
- | Qasper | Ask questions based on a single paper, questions proposed by NLP readers, and answered by NLP practitioners |
100
  | GovReport | A summarization task that requires summarizing government work reports |
101
  | QMSum | A summarization task that requires summarizing meeting records based on user queries |
102
  | VCSUM | A summarization task that requires summarizing Chinese meeting records |
103
- | TriviaQA | Single document question answering task, providing several Few Shot examples |
104
- | NQ | Single document question answering task, providing several Few Shot examples |
105
  | TREC | A classification task that requires categorizing questions, includes 50 categories in total |
106
  | LSHT | A Chinese classification task that requires categorizing news, includes 24 categories in total |
107
- | PassageRetrieval-en | Given 30 English Wikipedia paragraphs, determine which paragraph the given summary belongs to |
108
- | PassageCount | Determine the number of non-repeating paragraphs in a given number of paragraphs |
109
- | PassageRetrieval-zh | Given several Chinese paragraphs from the C4 data set, determine which paragraph the given abstract belongs to |
110
- | LCC | Given a longer piece of code, predict the next line of code |
111
- | RepoBench-P | Given code in multiple files within a GitHub repository (including inter-file dependencies), predict the next line of code |
112
 
113
 
114
  # Task construction
115
 
116
  > Note: For all tasks constructed from existing datasets, we use data from the validation or test set of the existing dataset (except for VCSUM).
117
 
118
- - The tasks of [HotpotQA](https://hotpotqa.github.io/), [2WikiMultihopQA](https://aclanthology.org/2020.coling-main.580/), [Musique](https://arxiv.org/abs/2108.00573), and [DuReader](https://github.com/baidu/DuReader) are built based on the original datasets and processed to make them suitable for long text evaluation. Specifically, for questions in the validation set, we select the evidence passage that contains the answer and several distracting articles. These articles together with the original question constitute the input of the related tasks.
119
- - The tasks of MultiFiedQA-zh and MultiFieldQA-en consist of long-text data from about 10 sources, including Latex papers, judicial documents, government work reports, and PDF documents indexed by Google. For each long text, we invite several PhD and master students to annotate, i.e., to ask questions based on the long text and give the correct answers. To better automate evaluation, we require the annotators to ask questions with definitive answers as much as possible.
120
  - The tasks of [NarrativeQA](https://arxiv.org/pdf/1712.07040.pdf), [Qasper](https://arxiv.org/pdf/2105.03011.pdf), [GovReport](https://arxiv.org/pdf/2104.02112.pdf), and [QMSum](https://arxiv.org/pdf/2104.05938.pdf) directly use the data provided by the original papers. In the specific construction, we use the template provided by [ZeroSCROLLS](https://www.zero.scrolls-benchmark.com/) to convert the corresponding data into pure text input.
121
- - The [VCSUM](https://arxiv.org/abs/2305.05280) task is built based on the original dataset, and we have designed a corresponding template to convert the corresponding data into pure text input.
122
  - The tasks of [TriviaQA](https://nlp.cs.washington.edu/triviaqa/) and [NQ](https://ai.google.com/research/NaturalQuestions/) are constructed in the manner of [CoLT5](https://arxiv.org/abs/2303.09752), which provides several examples of question and answering based on documents, and requires the language model to answer related questions based on new documents.
123
  - The tasks of [TREC](https://aclanthology.org/C02-1150.pdf) and [LSHT](http://tcci.ccf.org.cn/conference/2014/dldoc/evatask6.pdf) are built based on the original datasets. For each question in the validation set, we sample several data from the training set to form few-shot examples. These examples together with the questions in the validation set constitute the input for this task.
124
- - The PassageRetrieval-en task is constructed based on English Wikipedia. For each piece of data, we randomly sample 30 paragraphs from English Wikipedia and select one for summarization (using GPT3.5 Turbo). The task requires the model to specify which original paragraph the summary corresponds to.
125
- - The PassageCount task is constructed based on the English wiki. For each piece of data, we randomly sample several passages from English Wikipedia, repeat each paragraph at random several times, and finally shuffle the paragraphs at random. This task requires the model to determine how many paragraphs are not repeated in a given number of paragraphs.
126
- - The PasskeyRetrieval-zh task is constructed based on [C4](https://arxiv.org/abs/1910.10683). For each piece of data, we randomly sample several Chinese paragraphs from C4 and select one of them for summary (using GPT3.5 Turbo). This task requires the model to give the original paragraph to which the summary should correspond.
127
- - For the [LCC](https://arxiv.org/abs/2306.14893) task, we sample from the original code completion dataset. In the [RepoBench-P](https://arxiv.org/abs/2306.03091) task, we select the most challenging XF-F (Cross-File-First) setting from the original dataset and refer to the Oracle-Filled scenario in the original text. For each original piece of data, we randomly extract multiple cross-file code snippets, including the effective cross-file code snippet (gold snippet), and concatenate them as input, requiring the model to use the effective cross-file code to complete the current file's code.
 
22
 
23
  LongBench includes 13 English tasks, 5 Chinese tasks, and 2 code tasks, with the average length of most tasks ranging from 5k to 15k.
24
 
25
+ Github Repo for LongBench: https://github.com/THUDM/LongBench
26
+
27
  # How to use it?
28
 
29
  #### Loading Data
 
63
  # Task statistics
64
 
65
  | Task | Task Type | Eval metric | Avg len |Language | \#Sample |
66
+ | :-------- | :-----------:| :-----------: |:-------: | :-----------: |:--------: |
67
+ | HotpotQA | Multi-doc QA | F1 |9,149 |EN |200 |
68
+ | 2WikiMultihopQA| Multi-doc QA | F1 |4,885 |EN |200 |
69
+ | Musique| Multi-doc QA | F1 |7,798 |EN |200 |
70
+ | DuReader| Multi-doc QA | Rouge-L |15,768 |ZH |200 |
71
+ | MultiFieldQA-en| Single-doc QA | F1 |4,559 |EN |150 |
72
+ | MultiFieldQA-zh| Single-doc QA | F1 |6,771 |ZH |200 |
73
+ | NarrativeQA| Single-doc QA | F1 |18,405 |EN |200 |
74
+ | Qasper| Single-doc QA | F1 |3,619 |EN |200 |
75
+ | GovReport| Summarization | Rouge-L |8,169 |EN |200 |
76
+ | QMSum| Summarization | Rouge-L |10,546 |EN |200 |
77
+ | VCSUM| Summarization | Rouge-L |15,147 |ZH |200 |
78
+ | TriviaQA| Few shot | F1 |8,015 |EN |200 |
79
+ | NQ| Few shot | F1 |8,210 |EN |200 |
80
+ | TREC| Few shot | Accuracy |5,176 |EN |200 |
81
+ | LSHT| Few shot | Accuracy |22,333 |ZH |200 |
82
+ | PassageRetrieval-en| Synthetic | Accuracy |9,288 |EN |200 |
83
+ | PassageCount| Synthetic | Accuracy |11,141 |EN |200 |
84
+ | PassageRetrieval-zh | Synthetic | Accuracy |6,745 |ZH |200 |
85
+ | LCC| Code | Edit Sim |1,235 |Python/C#/Java |500 |
86
+ | RepoBench-P| Code | Edit Sim |5,622 |Python/Java |500 |
87
 
88
  > Note: In order to avoid discrepancies caused by different tokenizers, we use the word count (using Python's split function) to calculate the average length of English datasets and code datasets, and use the character count to calculate the average length of Chinese datasets.
89
 
90
  # Task description
91
 
92
  | Task | Task Description |
93
+ | :---------------- | :----------------------------------------------------------- |
94
  | HotpotQA | Answer related questions based on multiple given documents |
95
  | 2WikiMultihopQA | Answer related questions based on multiple given documents |
96
  | Musique | Answer related questions based on multiple given documents |
97
  | DuReader | Answer related Chinese questions based on multiple retrieved documents |
98
+ | MultiFieldQA-en | Answer English questions based on a long article, which comes from a relatively diverse field |
99
+ | MultiFieldQA-zh | Answer Chinese questions based on a long article, which comes from a relatively diverse field |
100
  | NarrativeQA | Ask questions based on stories or scripts, including understanding of important elements such as characters, plots, themes, etc. |
101
+ | Qasper | Ask questions based on a NLP research paper, questions proposed and answered by NLP practitioners |
102
  | GovReport | A summarization task that requires summarizing government work reports |
103
  | QMSum | A summarization task that requires summarizing meeting records based on user queries |
104
  | VCSUM | A summarization task that requires summarizing Chinese meeting records |
105
+ | TriviaQA | Single document question answering task, providing several few-shot examples |
106
+ | NQ | Single document question answering task, providing several few-shot examples |
107
  | TREC | A classification task that requires categorizing questions, includes 50 categories in total |
108
  | LSHT | A Chinese classification task that requires categorizing news, includes 24 categories in total |
109
+ | PassageRetrieval-en | Given 30 English Wikipedia paragraphs, determine which paragraph the given summary corresponds to |
110
+ | PassageCount | Determine the total number of different paragraphs in a given repetitive article |
111
+ | PassageRetrieval-zh | Given several Chinese paragraphs from the C4 data set, determine which paragraph the given abstract corresponds to |
112
+ | LCC | Given a long piece of code, predict the next line of code |
113
+ | RepoBench-P | Given code in multiple files within a GitHub repository (including cross-file dependencies), predict the next line of code |
114
 
115
 
116
  # Task construction
117
 
118
  > Note: For all tasks constructed from existing datasets, we use data from the validation or test set of the existing dataset (except for VCSUM).
119
 
120
+ - The tasks of [HotpotQA](https://hotpotqa.github.io/), [2WikiMultihopQA](https://aclanthology.org/2020.coling-main.580/), [Musique](https://arxiv.org/abs/2108.00573), and [DuReader](https://github.com/baidu/DuReader) are built based on the original datasets and processed to be suitable for long context evaluation. Specifically, for questions in the validation set, we select the evidence passage that contains the answer and several distracting articles. These articles together with the original question constitute the input of the tasks.
121
+ - The tasks of MultiFiedQA-zh and MultiFieldQA-en consist of long artical data from about 10 sources, including Latex papers, judicial documents, government work reports, and PDF documents indexed by Google. For each long artical, we invite several PhD and master students to annotate, i.e., to ask questions based on the long artical and give the correct answers. To better automate evaluation, we ask the annotators to propose questions with definitive answers as much as possible.
122
  - The tasks of [NarrativeQA](https://arxiv.org/pdf/1712.07040.pdf), [Qasper](https://arxiv.org/pdf/2105.03011.pdf), [GovReport](https://arxiv.org/pdf/2104.02112.pdf), and [QMSum](https://arxiv.org/pdf/2104.05938.pdf) directly use the data provided by the original papers. In the specific construction, we use the template provided by [ZeroSCROLLS](https://www.zero.scrolls-benchmark.com/) to convert the corresponding data into pure text input.
123
+ - The [VCSUM](https://arxiv.org/abs/2305.05280) task is built based on the original dataset, and we design a corresponding template to convert the corresponding data into pure text input.
124
  - The tasks of [TriviaQA](https://nlp.cs.washington.edu/triviaqa/) and [NQ](https://ai.google.com/research/NaturalQuestions/) are constructed in the manner of [CoLT5](https://arxiv.org/abs/2303.09752), which provides several examples of question and answering based on documents, and requires the language model to answer related questions based on new documents.
125
  - The tasks of [TREC](https://aclanthology.org/C02-1150.pdf) and [LSHT](http://tcci.ccf.org.cn/conference/2014/dldoc/evatask6.pdf) are built based on the original datasets. For each question in the validation set, we sample several data from the training set to form few-shot examples. These examples together with the questions in the validation set constitute the input for this task.
126
+ - The PassageRetrieval-en task is constructed based on English Wikipedia. For each piece of data, we randomly sample 30 paragraphs from English Wikipedia and select one for summarization (using GPT-3.5-Turbo). This task requires the model to give the original paragraph name to which the summary corresponds.
127
+ - The PassageCount task is constructed based on the English wiki. For each piece of data, we randomly sample several passages from English Wikipedia, repeat each paragraph at random several times, and finally shuffle the paragraphs. This task requires the model to determine the total number of different paragraphs in the given context.
128
+ - The PasskeyRetrieval-zh task is constructed based on [C4](https://arxiv.org/abs/1910.10683). For each piece of data, we randomly sample several Chinese paragraphs from C4 and select one of them for summarization (using GPT-3.5-Turbo). This task requires the model to give the original paragraph name to which the summary corresponds.
129
+ - For the [LCC](https://arxiv.org/abs/2306.14893) task, we sample from the original code completion dataset. In the [RepoBench-P](https://arxiv.org/abs/2306.03091) task, we select the most challenging XF-F (Cross-File-First) setting from the original dataset and refer to the Oracle-Filled scenario in the paper. For each original piece of data, we randomly extract multiple cross-file code snippets, including the gold cross-file code snippet, and concatenate them as input, requiring the model to effectively use cross-file code for completion.