LimRank: Less is More for Reasoning-Intensive Information Reranking

Model/Data Links | Experiments | Usage | Citation

Official repository for paper LimRank: Less is More for Reasoning-Intensive Information Reranking.

Links

Experiments

To reproduce the experiments, you can use the following code with uv for fast, reliable dependency management:

conda activate limrank_env
pip install -r requirements.txt

IR Experiments

Please refer to the Rerank Experiments for more details.

RAG Experiments

Please refer to the LimRank-GPQA for more details.

Citing

If you think our paper is useful, you can cite:

@misc{song2025limrankreasoningintensiveinformationreranking,
      title={LimRank: Less is More for Reasoning-Intensive Information Reranking}, 
      author={Tingyu Song and Yilun Zhao and Siyue Zhang and Chen Zhao and Arman Cohan},
      year={2025},
      eprint={2510.23544},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2510.23544}, 
}

Acknowledgements

We would like to thank the authors of the following papers and repos for their open-source contributions.

• rank1
• ReasonIR
• MTEB

License

MIT