用强化学习让大模型先推理再排序，仅用179样本就超越GPT-4。
REARANK: Reasoning Re-ranking Agent via Reinforcement Learning
- 先推理后重排，提升排序效果与可解释性。
- 仅需179个标注样本，在多个榜单上超越基线模型。
- 适合需要高精度排序与可解释性的信息检索场景。
我们提出REARANK，一个基于大语言模型（LLM）的列表式推理重排代理。REARANK在重排前显式进行推理，显著提升性能与可解释性。利用强化学习与数据增强，REARANK在多个主流信息检索基准上实现显著提升，尤其仅需179个标注样本。基于Qwen2.5-7B构建的REARANK-7B在域内与域外基准上表现接近GPT-4，甚至在推理密集型BRIGHT基准上超越GPT-4。结果验证了该方法的有效性，并表明强化学习能有效增强大模型在重排任务中的推理能力。
原文摘要 · Abstract (English)
We present REARANK, a large language model (LLM)-based listwise reasoning reranking agent. REARANK explicitly reasons before reranking, significantly improving both performance and interpretability. Leveraging reinforcement learning and data augmentation, REARANK achieves substantial improvements over baseline models across popular information retrieval benchmarks, notably requiring only 179 annotated samples. Built on top of Qwen2.5-7B, our REARANK-7B demonstrates performance comparable to GPT-4 on both in-domain and out-of-domain benchmarks and even surpasses GPT-4 on reasoning-intensive BRIGHT benchmarks. These results underscore the effectiveness of our approach and highlight how reinforcement learning can enhance LLM reasoning capabilities in reranking.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。