让大模型先选证据再生成，提升问答鲁棒性。
Aligning Extraction and Generation for Robust Retrieval-Augmented Generation
- 提取与生成联合优化，动态筛选相关文本并抑制噪声
- 在噪声检索下仍能生成准确答案，优于独立压缩模型
- 适合需要高可靠性问答的场景，如医疗、法律
检索增强生成（RAG）通过引入外部知识提升大模型能力，但生成过程易受检索噪声和相关片段定位不准影响，常导致幻觉。本文提出Ext2Gen框架，采用先提取后生成的设计，通过联合证据选择与答案生成，在动态识别查询相关文本的同时抑制噪声，无需独立的预生成压缩模块。该方法通过高质量成对反馈进行偏好对齐优化，在噪声或不精确检索条件下仍能生成准确且忠实的答案。实验表明，其显著增强了生成主干的鲁棒性，性能提升超过依赖独立压缩模型的方法（如Recomp、CompAct、EXIT）。此外，它还能从查询重写等改进的检索技术中获益，表明生成端优化可解决仅靠检索无法克服的局限。
原文摘要 · Abstract (English)
Retrieval-augmented generation (RAG) enhances LLMs with external knowledge, yet generation remains vulnerable to retrieval-induced noise and uncertain placement of relevant chunks, often causing hallucinations. We present Ext2Gen, an extract-then-generate framework that strengthens LLMs via joint evidence selection and answer generation, dynamically identifying query-relevant content while suppressing noise, thereby removing the need for any independent pre-generation compression module. Optimized through preference alignment with well-curated pairwise feedback, Ext2Gen produces accurate and faithful answers even under noisy or imprecise retrieval. Experiments demonstrate that it substantially enhances the robustness of the generation backbone and yields greater performance gains than methods relying on independent compression models, e.g., Recomp, CompAct, EXIT). It further benefits from improved retrieval techniques such as query rewriting, underscoring that generation-side enhancements address limitations that retrieval alone cannot overcome.
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。