arXiv 2601.164622026-05-31cs.CL

寻找关键：通过演化索引锚定上下文知识进行迭代检索

Finding What Matters: Anchoring Context Knowledge with Evolving Indices for Iterative Retrieval

School of Computer Science and Engineering, Northeastern University, China（东北大学计算机科学与工程学院）；Department of Computer Science and Technology, Institute for AI, Tsinghua University, China（清华大学人工智能研究院计算机科学与技术系）；Beijing National Research Center for Information Science and Technology, China（北京信息科学与技术国家研究中心）；School of Computer Science, Beijing University of Posts and Telecommunications, China（北京邮电大学计算机学院）

Mingyan Wu, Zhenghao Liu, Xinze Li, Yuqing Lan, Yukun Yan, Shuo Wang, Cheng Yang, Minghe Yu, Zheni Zeng, Maosong Sun

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

提出KAIR框架，通过迭代检索中动态更新的知识索引锚定关键证据，引导大语言模型在多跳问答中有效推理并缓解噪声干扰。

摘要：

检索增强生成（RAG）已成为通过引入外部知识来减轻大语言模型（LLMs）幻觉的主流范式。然而，现有的RAG系统通常难以有效整合和推理分散在嘈杂检索文档中的关键证据，尤其是在多跳场景中。本文提出KAIR，一种用于迭代检索的知识锚定框架，该框架在检索到的知识中锚定知识，以引导LLMs定位关键信息。在迭代检索过程中，KAIR逐步更新知识索引，以锚定检索文档中的显著证据。演化索引作为导航锚定索引，使LLM能够评估知识充分性并制定后续检索查询。最后，KAIR通过联合利用检索文档和最终确定的锚定索引生成答案。在四个多跳问答基准上的实验表明，KAIR始终优于强RAG基线。进一步分析显示，KAIR有效锚定关键知识并减轻了迭代检索过程中的上下文噪声，提高了LLM关联和推理分散在检索文档中的证据的能力。所有代码和数据可在https://github.com/NEUIR/KAIR获取。

英文摘要：

Retrieval-Augmented Generation (RAG) has become a dominant paradigm for mitigating hallucinations in Large Language Models (LLMs) by incorporating external knowledge. However, existing RAG systems often struggle to effectively integrate and reason over key evidence scattered across noisy retrieved documents, particularly in multi-hop scenarios. In this paper, we propose KAIR, a Knowledge Anchoring framework for Iterative Retrieval that anchors knowledge within retrieved knowledge to guide LLMs to locate the key information. During iterative retrieval, KAIR progressively updates the knowledge index to anchor salient evidence from retrieved documents. The evolving index serves as a navigational anchoring index that enables the LLM to assess knowledge sufficiency and formulate subsequent retrieval queries. Finally, KAIR generates answers by jointly leveraging the retrieved documents and the finalized anchoring index. Experiments on four multi-hop question answering benchmarks demonstrate that KAIR consistently outperforms strong RAG baselines. Further analysis shows that KAIR effectively anchors key knowledge and alleviates the context noise during iterative retrieval, improving the LLM's ability to associate and reason over dispersed evidence across retrieved documents. All code and data are available at https://github.com/NEUIR/KAIR.

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。