用户可借助该项目开展实时人机协作研究与实验,其核心是基于双加工理论构建语言智能体框架,支持单智能体、协作及人类实验,已在Overcooked Challenge等场景验证效果。【此简介由AI生成】
| 文件 | 最后提交记录 | 最后更新时间 |
|---|---|---|
| 1 年前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 10 个月前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 1 年前 |
以下内容由 AI 翻译,如有问题请 点此提交 issue 反馈
DPT-Agent
本文是论文《Leveraging Dual Process Theory in Language Agent Framework for Simultaneous Human-AI Collaboration》(已被 ACL 2025 主会接收)的官方实现。

📄 论文 | 🌐 网站 | 📘 机器之心 | 🧪 AGI-Eval
🔥最新动态
- [2025/05/16] DPT-agent 论文已被 ACL 2025 接收!
- [2025/03/18] 我们的研究成果被微信公众号机器之心报道!
- [2025/03/06] 我们已与AGI-Eval平台建立合作关系。Overcooked Challenge 的基准测试结果现已在AGI-Eval-Overcooked Challenge上公布。
Overcooked 挑战

DPT-Agent 框架

实验结果


使用说明
安装
创建新环境
conda create -n dptagent python=3.10 -y # support python<=3.10
conda activate dptagent
# Install pytorch
# We do not need gpu
pip install torch torchvision torchaudio
# Installing dependencies
bash ./install.sh
运行说明
设置大型语言模型
litellm -c llms/litellm/config.yml --port 40000
# check health
curl --location 'http://127.0.0.1:40000/health' -H "Authorization: Bearer sk-1234"
将LLM作为独立的系统1和系统2运行实验
对于单智能体实验,首先修改config/envs/overcooked.yaml
mode: burger_exp1
...
num_agents: 1
然后运行:
sh scripts/exp0/openai.sh
运行单智能体实验
对于单智能体实验,首先修改 config/envs/overcooked.yaml
mode: burger_exp1
...
num_agents: 1
然后运行:
sh scripts/exp1/openai.sh
使用基于规则的智能体运行协作实验
对于协作实验,首先修改 config/envs/overcooked.yaml
mode: burger_exp1
...
num_agents: 2
然后运行:
sh scripts/exp2/openai.sh
运行人工实验
如需使用地图 2,请修改 config/envs/overcooked.yaml
mode: burger_aa_new
...
num_agents: 2
然后运行:
sh scripts/overcooked/human_llm_app.sh
然后打开网站 http://localhost:5001
帮助
如需了解更多信息,请运行
python llm_agent_run_act.py --help
开发指南
我们建议在提交代码前使用 pre-commit 来统一格式。
# Init
pre-commit install
# You can run manually (maybe multiple times)
pre-commit run --all-files
# or it will automatically run when you try to commit, which is slow and seems stuck.
引用
@inproceedings{zhang-etal-2025-leveraging,
title = "Leveraging Dual Process Theory in Language Agent Framework for Real-time Simultaneous Human-{AI} Collaboration",
author = "Zhang, Shao and Wang, Xihuai and Zhang, Wenhao and Li, Chaoran and Song, Junru and Li, Tingyu and Qiu, Lin and Cao, Xuezhi and Cai, Xunliang and Yao, Wen and Zhang, Weinan and Wang, Xinbing and Wen, Ying",
editor = "Che, Wanxiang and Nabende, Joyce and Shutova, Ekaterina and Pilehvar, Mohammad Taher",
booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = jul,
year = "2025",
address = "Vienna, Austria",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.acl-long.206/",
doi = "10.18653/v1/2025.acl-long.206",
pages = "4081--4108",
ISBN = "979-8-89176-251-0",
}