NER

# Adapting Large Language Models to Domains via Continual Pre-Training
This repo contains the **NER dataset** used in our **ICLR 2024** paper [Adapting Large Language Models via Reading Comprehension](https://huggingface.co/papers/2309.09530).

We explore **continued pre-training on domain-specific corpora** for large language models. While this approach enriches LLMs with domain knowledge, it significantly hurts their prompting ability for question answering. Inspired by human learning via reading comprehension, we propose a simple method to **transform large-scale pre-training corpora into reading comprehension texts**, consistently improving prompting performance across tasks in biomedicine, finance, and law domains. **Our 7B model competes with much larger domain-specific models like BloombergGPT-50B**.

### [2024/11/29] 🤗 Introduce the multimodal version of AdaptLLM at [AdaMLLM](https://huggingface.co/AdaptLLM/Adapt-MLLM-to-Domains), for adapting MLLMs to domains 🤗

**************************** **Updates** ****************************
* 2024/11/29: Released [AdaMLLM](https://huggingface.co/AdaptLLM/Adapt-MLLM-to-Domains) for adapting MLLMs to domains
* 2024/9/20: Our [research paper for Instruction-Pretrain](https://huggingface.co/papers/2406.14491) has been accepted by EMNLP 2024
* 2024/8/29: Updated [guidelines](https://huggingface.co/datasets/AdaptLLM/finance-tasks) on evaluating any 🤗Huggingface models on the domain-specific tasks
* 2024/6/22: Released the [benchmarking code](https://github.com/microsoft/LMOps/tree/main/adaptllm)
* 2024/6/21: Released the general version of AdaptLLM at [Instruction-Pretrain](https://huggingface.co/instruction-pretrain)
* 2024/4/2: Released the [raw data splits (train and test)](https://huggingface.co/datasets/AdaptLLM/ConvFinQA) of all the evaluation datasets
* 2024/1/16: Our [research paper for AdaptLLM](https://huggingface.co/papers/2309.09530) has been accepted by ICLR 2024
* 2023/12/19: Released our [13B base models](https://huggingface.co/AdaptLLM/law-LLM-13B) developed from LLaMA-1-13B
* 2023/12/8: Released our [chat models](https://huggingface.co/AdaptLLM/law-chat) developed from LLaMA-2-Chat-7B
* 2023/9/18: Released our [paper](https://huggingface.co/papers/2309.09530), [code](https://github.com/microsoft/LMOps), [data](https://huggingface.co/datasets/AdaptLLM/law-tasks), and [base models](https://huggingface.co/AdaptLLM/law-LLM) developed from LLaMA-1-7B

## Domain-Specific LLaMA-1
### LLaMA-1-7B
In our paper, we develop three domain-specific models from LLaMA-1-7B, which are also available in Huggingface: [Biomedicine-LLM](https://huggingface.co/AdaptLLM/medicine-LLM), [Finance-LLM](https://huggingface.co/AdaptLLM/finance-LLM) and [Law-LLM](https://huggingface.co/AdaptLLM/law-LLM), the performances of our AdaptLLM compared to other domain-specific LLMs are:

<p align='center'>
<img src="https://cdn-uploads.huggingface.co/production/uploads/650801ced5578ef7e20b33d4/6efPwitFgy-pLTzvccdcP.png" width="700">
</p>

### LLaMA-1-13B
Moreover, we scale up our base model to LLaMA-1-13B to see if **our method is similarly effective for larger-scale models**, and the results are consistently positive too: [Biomedicine-LLM-13B](https://huggingface.co/AdaptLLM/medicine-LLM-13B), [Finance-LLM-13B](https://huggingface.co/AdaptLLM/finance-LLM-13B) and [Law-LLM-13B](https://huggingface.co/AdaptLLM/law-LLM-13B).

## Domain-Specific LLaMA-2-Chat
Our method is also effective for aligned models! LLaMA-2-Chat requires a [specific data format](https://huggingface.co/blog/llama2#how-to-prompt-llama-2), and our **reading comprehension can perfectly fit the data format** by transforming the reading comprehension into a multi-turn conversation. We have also open-sourced chat models in different domains: [Biomedicine-Chat](https://huggingface.co/AdaptLLM/medicine-chat), [Finance-Chat](https://huggingface.co/AdaptLLM/finance-chat) and [Law-Chat](https://huggingface.co/AdaptLLM/law-chat)

## Domain-Specific Tasks

### Pre-templatized/Formatted Testing Splits
To easily reproduce our prompting results, we have uploaded the filled-in zero/few-shot input instructions and output completions of the test each domain-specific task: [biomedicine-tasks](https://huggingface.co/datasets/AdaptLLM/medicine-tasks), [finance-tasks](https://huggingface.co/datasets/AdaptLLM/finance-tasks), and [law-tasks](https://huggingface.co/datasets/AdaptLLM/law-tasks).

**Note:** those filled-in instructions are specifically tailored for models before alignment and do NOT fit for the specific data format required for chat models.

### Raw Datasets
We have also uploaded the raw training and testing splits, for facilitating fine-tuning or other usages:
- [ChemProt](https://huggingface.co/datasets/AdaptLLM/ChemProt)
- [RCT](https://huggingface.co/datasets/AdaptLLM/RCT)
- [ConvFinQA](https://huggingface.co/datasets/AdaptLLM/ConvFinQA)
- [FiQA_SA](https://huggingface.co/datasets/AdaptLLM/FiQA_SA)
- [Headline](https://huggingface.co/datasets/AdaptLLM/Headline)
- [NER](https://huggingface.co/datasets/AdaptLLM/NER)
- [FPB](https://huggingface.co/datasets/AdaptLLM/FPB)

The other datasets used in our paper have already been available in huggingface, and you can directly load them with the following code:
```python
from datasets import load_dataset

# MQP:
dataset = load_dataset('medical_questions_pairs')
# PubmedQA:
dataset = load_dataset('bigbio/pubmed_qa')
# USMLE:
dataset=load_dataset('GBaker/MedQA-USMLE-4-options')
# SCOTUS
dataset = load_dataset("lex_glue", 'scotus')
# CaseHOLD
dataset = load_dataset("lex_glue", 'case_hold')
# UNFAIR-ToS
dataset = load_dataset("lex_glue", 'unfair_tos')
```

## Citation
If you find our work helpful, please cite us:
```bibtex
@inproceedings{
cheng2024adapting,
title={Adapting Large Language Models via Reading Comprehension},
author={Daixuan Cheng and Shaohan Huang and Furu Wei},
booktitle={The Twelfth International Conference on Learning Representations},
year={2024},
url={https://openreview.net/forum?id=y886UXPEZ0}
}
```

and the original dataset:
```bibtex
@inproceedings{NER,
author = {Julio Cesar Salinas Alvarado and
Karin Verspoor and
Timothy Baldwin},
title = {Domain Adaption of Named Entity Recognition to Support Credit Risk
Assessment},
booktitle = {{ALTA}},
pages = {84--90},
publisher = {{ACL}},
year = {2015}
}
```

# 基于持续预训练的大语言模型领域适配
本仓库包含我们发表于**国际学习表征会议2024（ICLR 2024）**的论文《基于阅读理解的大语言模型适配》中所使用的**命名实体识别（Named Entity Recognition，NER）数据集**，论文链接为：https://huggingface.co/papers/2309.09530。

我们针对大语言模型（Large Language Model，LLM）探索了**领域专属语料上的持续预训练**方案。尽管该方法可为LLM注入领域知识，但会显著损害其问答任务的提示学习能力。受人类通过阅读理解进行学习的启发，我们提出了一种简单方法，可将大规模预训练语料转换为阅读理解文本，该方法能持续提升生物医学、金融、法律三大领域各类任务的提示学习性能。**我们的7B模型可与BloombergGPT-50B等超大规模领域专属模型相媲美**。

### [2024/11/29] 🤗 推出AdaptLLM的多模态版本[AdaMLLM](https://huggingface.co/AdaptLLM/Adapt-MLLM-to-Domains)，用于多模态大语言模型（Multimodal Large Language Model，MLLM）的领域适配 🤗

**************************** **更新日志** ****************************
* 2024/11/29：发布用于多模态大语言模型领域适配的[AdaMLLM](https://huggingface.co/AdaptLLM/Adapt-MLLM-to-Domains)
* 2024/09/20：我们的[指令预训练相关研究论文](https://huggingface.co/papers/2406.14491)被自然语言处理经验方法会议2024（EMNLP 2024）收录
* 2024/08/29：更新了针对任意🤗Huggingface模型的领域专属任务评估[指南](https://huggingface.co/datasets/AdaptLLM/finance-tasks)
* 2024/06/22：发布了基准测试[代码](https://github.com/microsoft/LMOps/tree/main/adaptllm)
* 2024/06/21：在[指令预训练](https://huggingface.co/instruction-pretrain)平台发布AdaptLLM通用版本
* 2024/04/02：发布了所有评估数据集的[原始训练与测试划分数据](https://huggingface.co/datasets/AdaptLLM/ConvFinQA)
* 2024/01/16：我们的[AdaptLLM相关研究论文](https://huggingface.co/papers/2309.09530)被国际学习表征会议2024（ICLR 2024）收录
* 2023/12/19：发布基于LLaMA-1-13B开发的[13B基座模型](https://huggingface.co/AdaptLLM/law-LLM-13B)
* 2023/12/08：发布基于LLaMA-2-Chat-7B开发的[对话模型](https://huggingface.co/AdaptLLM/law-chat)
* 2023/09/18：发布了基于LLaMA-1-7B开发的[研究论文](https://huggingface.co/papers/2309.09530)、[代码](https://github.com/microsoft/LMOps)、[数据集](https://huggingface.co/datasets/AdaptLLM/law-tasks)以及[基座模型](https://huggingface.co/AdaptLLM/law-LLM)

## 领域专属LLaMA-1模型
### LLaMA-1-7B
在本研究中，我们基于LLaMA-1-7B开发了三款领域专属模型，现已上架Huggingface平台：[生物医学-LLM](https://huggingface.co/AdaptLLM/medicine-LLM)、[金融-LLM](https://huggingface.co/AdaptLLM/finance-LLM)以及[法律-LLM](https://huggingface.co/AdaptLLM/law-LLM)。本研究提出的AdaptLLM与其他领域专属大语言模型的性能对比如下：

<p align='center'>
<img src="https://cdn-uploads.huggingface.co/production/uploads/650801ced5578ef7e20b33d4/6efPwitFgy-pLTzvccdcP.png" width="700">
</p>

### LLaMA-1-13B
此外，我们将基座模型扩展至LLaMA-1-13B，以验证**我们的方法对更大规模模型同样有效**，实验结果同样表现优异：[生物医学-LLM-13B](https://huggingface.co/AdaptLLM/medicine-LLM-13B)、[金融-LLM-13B](https://huggingface.co/AdaptLLM/finance-LLM-13B)以及[法律-LLM-13B](https://huggingface.co/AdaptLLM/law-LLM-13B)。

## 领域专属LLaMA-2-Chat模型
我们的方法对对齐后的模型同样有效！LLaMA-2-Chat需要遵循[特定数据格式](https://huggingface.co/blog/llama2#how-to-prompt-llama-2)，我们可将阅读理解文本转换为多轮对话，完美适配该数据格式。我们还开源了多个领域的对话模型：[生物医学-Chat](https://huggingface.co/AdaptLLM/medicine-chat)、[金融-Chat](https://huggingface.co/AdaptLLM/finance-chat)以及[法律-Chat](https://huggingface.co/AdaptLLM/law-chat)

## 领域专属任务
### 预模板化/格式化测试集划分
为便于复现我们的提示学习实验结果，我们上传了各领域专属任务的零样本/少样本输入提示与测试集输出补全的填充示例：[生物医学任务集](https://huggingface.co/datasets/AdaptLLM/medicine-tasks)、[金融任务集](https://huggingface.co/datasets/AdaptLLM/finance-tasks)以及[法律任务集](https://huggingface.co/datasets/AdaptLLM/law-tasks)。**注意：** 这些填充示例专为对齐前的模型设计，不适用于对话模型所需的特定数据格式。

### 原始数据集
我们还上传了各任务的原始训练与测试集划分，以支持微调或其他应用场景：
- [ChemProt](https://huggingface.co/datasets/AdaptLLM/ChemProt)
- [RCT](https://huggingface.co/datasets/AdaptLLM/RCT)
- [ConvFinQA](https://huggingface.co/datasets/AdaptLLM/ConvFinQA)
- [FiQA_SA](https://huggingface.co/datasets/AdaptLLM/FiQA_SA)
- [Headline](https://huggingface.co/datasets/AdaptLLM/Headline)
- [NER](https://huggingface.co/datasets/AdaptLLM/NER)
- [FPB](https://huggingface.co/datasets/AdaptLLM/FPB)

本论文中使用的其他数据集已上架Huggingface平台，您可通过以下代码直接加载：
python
from datasets import load_dataset

# MQP:
dataset = load_dataset('medical_questions_pairs')
# PubmedQA:
dataset = load_dataset('bigbio/pubmed_qa')
# USMLE:
dataset=load_dataset('GBaker/MedQA-USMLE-4-options')
# SCOTUS
dataset = load_dataset("lex_glue", 'scotus')
# CaseHOLD
dataset = load_dataset("lex_glue", 'case_hold')
# UNFAIR-ToS
dataset = load_dataset("lex_glue", 'unfair_tos')

## 引用声明
如果您的工作受益于本项目，请引用以下文献：
bibtex
@inproceedings{
cheng2024adapting,
title={Adapting Large Language Models via Reading Comprehension},
author={Daixuan Cheng and Shaohan Huang and Furu Wei},
booktitle={The Twelfth International Conference on Learning Representations},
year={2024},
url={https://openreview.net/forum?id=y886UXPEZ0}
}

以及原始数据集的引用：
bibtex
@inproceedings{NER,
author = {Julio Cesar Salinas Alvarado and
Karin Verspoor and
Timothy Baldwin},
title = {Domain Adaption of Named Entity Recognition to Support Credit Risk
Assessment},
booktitle = {{ALTA}},
pages = {84--90},
publisher = {{ACL}},
year = {2015}
}