DrNicefellow
commited on
Commit
β’
1de71b6
1
Parent(s):
7cfabb5
Update README.md
Browse files
README.md
CHANGED
@@ -8,68 +8,4 @@ pipeline_tag: text-generation
|
|
8 |
tags:
|
9 |
- llm-agent
|
10 |
---
|
11 |
-
|
12 |
-
<h1 align="center"> Executable Code Actions Elicit Better LLM Agents </h1>
|
13 |
-
|
14 |
-
<p align="center">
|
15 |
-
<a href="https://github.com/xingyaoww/code-act">π» Code</a>
|
16 |
-
β’
|
17 |
-
<a href="https://arxiv.org/abs/2402.01030">π Paper</a>
|
18 |
-
β’
|
19 |
-
<a href="https://huggingface.co/datasets/xingyaoww/code-act" >π€ Data (CodeActInstruct)</a>
|
20 |
-
β’
|
21 |
-
<a href="https://huggingface.co/xingyaoww/CodeActAgent-Mistral-7b-v0.1" >π€ Model (CodeActAgent-Mistral-7b-v0.1)</a>
|
22 |
-
β’
|
23 |
-
<a href="https://chat.xwang.dev/">π€ Chat with CodeActAgent!</a>
|
24 |
-
</p>
|
25 |
-
|
26 |
-
We propose to use executable Python **code** to consolidate LLM agentsβ **act**ions into a unified action space (**CodeAct**).
|
27 |
-
Integrated with a Python interpreter, CodeAct can execute code actions and dynamically revise prior actions or emit new actions upon new observations (e.g., code execution results) through multi-turn interactions.
|
28 |
-
|
29 |
-
![Overview](https://github.com/xingyaoww/code-act/blob/main/figures/overview.png?raw=true)
|
30 |
-
|
31 |
-
## Why CodeAct?
|
32 |
-
|
33 |
-
Our extensive analysis of 17 LLMs on API-Bank and a newly curated benchmark [M<sup>3</sup>ToolEval](docs/EVALUATION.md) shows that CodeAct outperforms widely used alternatives like Text and JSON (up to 20% higher success rate). Please check our paper for more detailed analysis!
|
34 |
-
|
35 |
-
![Comparison between CodeAct and Text/JSON](https://github.com/xingyaoww/code-act/blob/main/figures/codeact-comparison-table.png?raw=true)
|
36 |
-
*Comparison between CodeAct and Text / JSON as action.*
|
37 |
-
|
38 |
-
![Comparison between CodeAct and Text/JSON](https://github.com/xingyaoww/code-act/blob/main/figures/codeact-comparison-perf.png?raw=true)
|
39 |
-
*Quantitative results comparing CodeAct and {Text, JSON} on M<sup>3</sup>ToolEval.*
|
40 |
-
|
41 |
-
|
42 |
-
|
43 |
-
## π CodeActInstruct
|
44 |
-
|
45 |
-
We collect an instruction-tuning dataset CodeActInstruct that consists of 7k multi-turn interactions using CodeAct. Dataset is release at [huggingface dataset π€](https://huggingface.co/datasets/xingyaoww/code-act). Please refer to the paper and [this section](#-data-generation-optional) for details of data collection.
|
46 |
-
|
47 |
-
|
48 |
-
![Data Statistics](https://github.com/xingyaoww/code-act/blob/main/figures/data-stats.png?raw=true)
|
49 |
-
*Dataset Statistics. Token statistics are computed using Llama-2 tokenizer.*
|
50 |
-
|
51 |
-
## πͺ CodeActAgent
|
52 |
-
|
53 |
-
Trained on **CodeActInstruct** and general conversaions, **CodeActAgent** excels at out-of-domain agent tasks compared to open-source models of the same size, while not sacrificing generic performance (e.g., knowledge, dialog). We release two variants of CodeActAgent:
|
54 |
-
- **CodeActAgent-Mistral-7b-v0.1** (recommended, [model link](https://huggingface.co/xingyaoww/CodeActAgent-Mistral-7b-v0.1)): using Mistral-7b-v0.1 as the base model with 32k context window.
|
55 |
-
- **CodeActAgent-Llama-7b** ([model link](https://huggingface.co/xingyaoww/CodeActAgent-Llama-2-7b)): using Llama-2-7b as the base model with 4k context window.
|
56 |
-
|
57 |
-
![Model Performance](https://github.com/xingyaoww/code-act/blob/main/figures/model-performance.png?raw=true)
|
58 |
-
*Evaluation results for CodeActAgent. ID and OD stand for in-domain and out-of-domain evaluation correspondingly. Overall averaged performance normalizes the MT-Bench score to be consistent with other tasks and excludes in-domain tasks for fair comparison.*
|
59 |
-
|
60 |
-
|
61 |
-
Please check out [our paper](TODO) and [code](https://github.com/xingyaoww/code-act) for more details about data collection, model training, and evaluation.
|
62 |
-
|
63 |
-
|
64 |
-
## π Citation
|
65 |
-
|
66 |
-
```bibtex
|
67 |
-
@misc{wang2024executable,
|
68 |
-
title={Executable Code Actions Elicit Better LLM Agents},
|
69 |
-
author={Xingyao Wang and Yangyi Chen and Lifan Yuan and Yizhe Zhang and Yunzhu Li and Hao Peng and Heng Ji},
|
70 |
-
year={2024},
|
71 |
-
eprint={2402.01030},
|
72 |
-
archivePrefix={arXiv},
|
73 |
-
primaryClass={cs.CL}
|
74 |
-
}
|
75 |
-
```
|
|
|
8 |
tags:
|
9 |
- llm-agent
|
10 |
---
|
11 |
+
This is a 6.0bpw h6 quantized version of [xingyaoww/CodeActAgent-Mistral-7b-v0.1](https://huggingface.co/xingyaoww/CodeActAgent-Mistral-7b-v0.1). It is quantized with exllamav2.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|