arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2604.06647cs.CL
收藏

基于反馈的检索增强生成反馈适应

Feedback Adaptation for Retrieval-Augmented Generation

Qualcomm AI Research(高通人工智能研究)

Jihwan Bang, Seunghan Yang, Kyuhong Shim, Simyung Chang, Juntae Lee, Sungha Choi

AI总结:

研究提出反馈适应问题,通过引入修正延迟和反馈后性能两个指标,评估RAG系统在反馈后的适应能力,并提出PatchRAG模型实现即时修正与强泛化。

摘要:

检索增强生成(RAG)系统通常在静态假设下评估,尽管在部署中常通过用户或专家反馈进行修正。现有评估协议关注整体准确率,未能捕捉反馈引入后系统适应性变化。本文引入反馈适应作为RAG系统的评估问题,探讨反馈如何有效且快速传播至未来查询。为使该行为可测量,我们提出两个评估轴:修正延迟,捕捉反馈提供与行为改变之间的延迟;反馈后性能,测量反馈后对语义相关查询的可靠性。使用这些指标,我们显示基于训练的方法在延迟修正与可靠适应之间存在权衡。进一步提出PatchRAG,一种最小化推理时间的实例,无需重新训练即可整合反馈,展示即时修正和强反馈后泛化能力。结果强调反馈适应是交互环境下RAG系统行为之前未被重视的维度。

英文摘要:

Retrieval-Augmented Generation (RAG) systems are typically evaluated under static assumptions, despite being frequently corrected through user or expert feedback in deployment. Existing evaluation protocols focus on overall accuracy and fail to capture how systems adapt after feedback is introduced. We introduce feedback adaptation as a problem setting for RAG systems, which asks how effectively and how quickly corrective feedback propagates to future queries. To make this behavior measurable, we propose two evaluation axes: correction lag, which captures the delay between feedback provision and behavioral change, and post-feedback performance, which measures reliability on semantically related queries after feedback. Using these metrics, we show that training-based approaches exhibit a trade-off between delayed correction and reliable adaptation. We further propose PatchRAG, a minimal inference-time instantiation that incorporates feedback without retraining, demonstrating immediate correction and strong post-feedback generalization under the proposed evaluation. Our results highlight feedback adaptation as a previously overlooked dimension of RAG system behavior in interactive settings.

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。

↑