让大模型在干扰信息中精准更新知识，一次通过即可完成
Robust Knowledge Editing via Explicit Reasoning Chains for Distractor-Resilient Multi-Hop QA
- 用四阶段推理链引导模型识别真实信息，过滤无关干扰
- 在含最多4个干扰项的多跳问答中准确率达90.2%
- 适合需要高可靠知识更新的场景，如医疗、金融领域
大型语言模型虽蕴含丰富世界知识，但训练后无法动态更新，全量重训成本过高。知识编辑技术应运而生，可在不重新训练的前提下注入或覆盖特定事实，但现有方法要么依赖表面线索，要么需复杂迭代流程，在噪声和多跳任务下易失效。本文提出Reason-KE，一种基于显式推理链的端到端编辑框架，通过四个结构化阶段——事实确认、相关性判断、选择性应用与最终推理——在单次通过中实现干扰过滤。在包含最多四个无关事实的MQuAKE-CF数据集上，该方法使Qwen2.5-7B在多跳问答中的准确率提升至90.2%，在强干扰下仅下降6.3%，答案泄露时下降不足1%。定量分析表明，Reason-KE具备优异的鲁棒性与效率，为可靠的知识更新建立了新基准。
原文摘要 · Abstract (English)
Large language models (LLMs) encode vast amounts of world knowledge but remain static once trained, making the timely integration of emerging facts prohibitively expensive via full retraining. Knowledge-editing techniques have thus emerged to inject or overwrite specific facts into LLMs, yet they either over-rely on superficial cues or incur complex, iterative pipelines that collapse under noisy, multi-hop conditions. We introduce Reason-KE, an end-to-end reasoning-chain-based editing framework that steers a pretrained LLM through four structured stages-fact acknowledgment, relevance determination, selective application, and final reasoning-to filter distractors in a single pass. Trained on MQuAKE-CF with up to four irrelevant facts, Reason-KE elevates Qwen2.5-7B's multi-hop QA accuracy to 90.2% while suffering merely a 6.3% drop under heavy distraction and <1% when answers are leaked. Our quantitative analysis confirms Reason-KE's resilience and efficiency, establishing a new state-of-the-art for reliable LLM knowledge updates.
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。