PrismRAG：借助干扰项韧性与策略化推理提升 RAG 事实性
PrismRAG: Boosting RAG Factuality with Distractor Resilience and Strategized Reasoning
AI总结：
针对 RAG 易受半相关干扰段落影响且推理不足的问题，本文提出 PrismRAG 微调框架，通过干扰项感知问答对和推理中心训练，在 12 个基准上平均提升事实性 5.4%。
摘要：
当检索到的上下文包含令人困惑的半相关段落，或回答问题需要深入的上下文理解与推理时，检索增强生成（RAG）往往表现不佳。我们提出一个名为 PrismRAG 的高效微调框架，它：（i）使用将黄金证据与细微干扰段落混合的、具备干扰项意识的问答对训练模型；（ii）灌输以推理为中心的习惯，使 LLM 能够规划、解释并综合，而无需依赖大量人工设计的指令。在涵盖不同应用领域和场景的 12 个开卷 RAG 问答基准上评估，PrismRAG 将平均事实性提升 5.4%，优于最先进的解决方案。
英文摘要：
Retrieval-augmented generation (RAG) often falls short when retrieved context includes confusing semi-relevant passages, or when answering questions require deep contextual understanding and reasoning. We propose an efficient fine-tuning framework, called PrismRAG, that (i) trains the model with distractor-aware QA pairs mixing gold evidence with subtle distractor passages, and (ii) instills reasoning-centric habits that make the LLM plan, rationalize, and synthesize without relying on extensive human engineered instructions. Evaluated across 12 open-book RAG QA benchmarks spanning diverse application domains and scenarios, PrismRAG improves average factuality by 5.4%, outperforming state-of-the-art solutions.
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。
