ChatCoT

KO | EN

gitlite — search

#hacktoberfest #react #ai #llm #go #android #machine-learning #deep-learning #linux #docker #macos #java #cli #awesome-list

(★ 46)

The official repository of "ChatCoT: Tool-Augmented Chain-of-Thought Reasoning on Chat-based Large Language Models"

ChatCoT 최신버젼 다운로드

최종 버전 다운로드 (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# ChatCoT: Tool-Augmented Chain-of-Thought Reasoning on Chat-based Large Language Models

This repo provides the source code & data of our paper: [ChatCoT: Tool-Augmented Chain-of-Thought Reasoning on Chat-based Large Language Models](https://arxiv.org/abs/2305.14323) (Arxiv 2023).

![](https://raw.githubusercontent.com/RUCAIBox/ChatCoT/main/picture/framework.png)

```
@InProceedings{Chen-ChatCot-2023,
      title = {ChatCoT: Tool-Augmented Chain-of-Thought Reasoning on Chat-based Large Language Models}, 
      author = {Zhipeng Chen and Kun Zhou and Beichen Zhang and Zheng Gong and Wayne Xin Zhao and Ji-Rong Wen},
      year = {2023},
      eprint = {2305.14323},
      archivePrefix = {arXiv},
      primaryClass = {cs.CL}
}
```

## Data & Code

+ `math/`: the code about ChatCoT on MATH dataset
    + `demo/`: the demos used in in-context learning on few-shot setting
    + `math/result/`: the result files of different methods
    + `math/scripts/`: the running scripts of different methods
    + `math/ablation`: the code of ablation study
    + `math/self_consistency`: the code to explore combining ChatCoT with CoT improvement strategies

The `hotpotqa/` folder is similar with `math/` folder

## Usage

### Prepare

You can use following scripts to install related python package through pip:
```
git clone https://github.com/RUCAIBox/ChatCoT.git
cd ChatCoT
pip install -r requirements.txt
```

### Inference

You can run ChatCot on the sub-task of MATH dataset by running `run_turbo_chatcot.sh`:
```
cd math
bash scripts/run_turbo_chatcot.sh
```

You have to replace `YOUR_API_KEY ` with you openai api key in the code. Specially, we run ChatCoT through multi-processing, and you should prepare a list of api key in order to run the code correctly.

### Evaluate

You can evaluate the results by running `eval.sh`:
```
cd math
bash scripts/eval.sh
```

## Results

### Main Results

| Methods | Algebra | CP | PC | PA | Geometry | IA | NT |
| :-----: | :-: | :-: | :-: |:-: | :-: | :-: |:-: |
| CoT | 48.10 | 31.43 | 21.06 | 56.60 | 22.34 | 18.27 | 29.07 |
| CoT w/ Tool | 35.89 | 22.57 | 9.34 | 40.53 | 13.57 | 9.41 | 19.44 |
| CoT w/ Retri | 52.74 | 32.70 | 18.86 | 58.44 | 29.23 | **19.93** | 31.67 |
| ChatCoT | 56.11 | **34.18** | **23.81** | **59.24** | **29.85** | 19.49 | **32.59** |

| Methods | HotpotQA |
| :-----: | :-: |
| CoT | 37.99 |
| CoT w/ Tool | 31.42 |
| ChatCoT w/o Feedback | 53.79 |
| ChatCoT | **59.16** |

### Ablation Study

| Methods | PC  | Geo | NT  |
| :-----: | :-: | :-: | :-: |
| ChatCoT | **23.81** | **29.85** | **32.59** |
| ChatCoT w/o TK | 23.26 | 29.23 | 30.56 |
| ChatCoT w/o RATK | 19.96 | 27.35 | 30.93 |
| ChatCoT w/o MRF | 21.61 | 24.22 | 32.22 |

The results of ablation study. TK, RATK, and MRF denote if using tool knowledge, retrieval-augmented task knowledge, and multi-turn reasoning format at early turns of the conversation, respectively.

### Combining CoT Improvement Strategies

| Methods | CP  | NT  |
| :-----: | :-: | :-: |
| CoT | 31.43 | 29.07 |
| CoT + SC | 35.23 | 34.44 |
| ChatCoT | 34.18 | 32.59 |
| ChatCoT + SC | **40.08** | **38.33** |

ChatCoT

ChatCoT 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

ChatCoT

ChatCoT 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Download README