arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.03689cs.CLcs.AI
收藏

重新思考重排序器:面向鲁棒检索增强生成的边界感知证据选择

Rethinking the Reranker: Boundary-Aware Evidence Selection for Robust Retrieval-Augmented Generation

University of Illinois Urbana-Champaign;Hong Kong University of Science;Washington University in St. Louis;Texas A\&M University
Jiashuo Sun, Pengcheng Jiang, Saizhuo Wang, Jiajun Fan, Heng Wang, Siru Ouyang, Ming Zhong, Yizhu Jiao, Chengsong Huang, Xueqiang Xu, Pengrui Han, Peiran Li, Ji… 展开作者

Jiashuo Sun, Pengcheng Jiang, Saizhuo Wang, Jiajun Fan, Heng Wang, Siru Ouyang, Ming Zhong, Yizhu Jiao, Chengsong Huang, Xueqiang Xu, Pengrui Han, Peiran Li, Jiaxin Huang, Ge Liu, Heng Ji, Jiawei Han

AI总结:

BAR-RAG通过边界感知证据选择提升检索增强生成在噪声下的鲁棒性与性能

摘要:

检索增强生成(RAG)系统在现实中的检索噪声下仍然脆弱,即使所需证据出现在前K结果中。关键原因在于检索器和重排序器仅优化相关性,往往选择要么是答案揭示性段落,要么缺乏回答问题所需关键信息的证据,而不考虑证据是否适合生成器。我们提出了BAR-RAG,将重排序器重新框架为一个边界感知的证据选择器,目标是生成器的Goldilocks Zone——证据既不 trivially 容易也不 fundamentally 无法回答,但具有挑战性却足够用于推理,从而提供最强的学习信号。BAR-RAG通过生成器反馈使用强化学习训练选择器,并采用两阶段流程,在诱导的证据分布下微调生成器以缓解训练和推理之间的分布不匹配。在知识密集型问答基准测试中,实验表明BAR-RAG在噪声检索下一致提升了端到端性能,相比强大的RAG和重排序基线平均提升10.3个百分点,同时显著提高了鲁棒性。代码可在https://github.com/GasolSun36/BAR-RAG公开获取。

英文摘要:

Retrieval-Augmented Generation (RAG) systems remain brittle under realistic retrieval noise, even when the required evidence appears in the top-K results. A key reason is that retrievers and rerankers optimize solely for relevance, often selecting either trivial, answer-revealing passages or evidence that lacks the critical information required to answer the question, without considering whether the evidence is suitable for the generator. We propose BAR-RAG, which reframes the reranker as a boundary-aware evidence selector that targets the generator's Goldilocks Zone -- evidence that is neither trivially easy nor fundamentally unanswerable for the generator, but is challenging yet sufficient for inference and thus provides the strongest learning signal. BAR-RAG trains the selector with reinforcement learning using generator feedback, and adopts a two-stage pipeline that fine-tunes the generator under the induced evidence distribution to mitigate the distribution mismatch between training and inference. Experiments on knowledge-intensive question answering benchmarks show that BAR-RAG consistently improves end-to-end performance under noisy retrieval, achieving an average gain of 10.3 percent over strong RAG and reranking baselines while substantially improving robustness. Code is publicly avaliable at https://github.com/GasolSun36/BAR-RAG.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑