metadata
license: mit
ARENA: Adaptive-Rewarded Evidence Navigation Agent
This is the official model release from our paper:
Effective and Transparent RAG: Adaptive-Reward Reinforcement Learning for Decision Traceability
This model is part of the ARENA framework, which improves the reasoning ability and interpretability of retrieval-augmented generation (RAG) by reinforcement learning with adaptive rewards.
For instructions on how to use the model and more implementation details, please refer to our GitHub repository:
👉 https://github.com/ren258/ARENA
Citation
If you find this work useful, please consider citing our paper:
@article{ren2025effective,
title={Effective and Transparent RAG: Adaptive-Reward Reinforcement Learning for Decision Traceability},
author={Ren, Jingyi and Xu, Yekun and Wang, Xiaolong and Li, Weitao and Ma, Weizhi and Liu, Yang},
journal={arXiv preprint arXiv:2505.13258},
year={2025}
}
Feel free to reach out via GitHub issues if you encounter any problems or have questions!