collabllm 0.0.3

(Oral @ ICML 2025) CollabLLM: From Passive Responders to Active Collaborators

CollabLLM: From Passive Responders to Active Collaborators (ICML 2025 Oral)

Paper

Installation

conda create -n collabllm python=3.10
pip install collabllm

You should further install additional packages for customized metric, such as bigcodebench. To further conduct trainining, you should install additional packages such as trl,

To reproduce our experiments:

conda create -n collabllm python=3.10
pip install -r requirements.txt

Quick Start

notebook_tutorials/

scripts/

Citation

If you use this code in your research, please cite the following paper:

@inproceedings{
    collabllm,
    title={CollabLLM: From Passive Responders to Active Collaborators},
    author={Shirley Wu and Michel Galley and 
            Baolin Peng and Hao Cheng and 
            Gavin Li and Yao Dou and Weixin Cai and 
            James Zou and Jure Leskovec and Jianfeng Gao
            },
    booktitle={ICML},
    year={2025}
}