CLAP: 共指链接增强的段落检索
CLAP: Coreference-Linked Augmentation for Passage Retrieval
AI总结：
针对基于LLM的段落扩展存在语义漂移、共指连续性被破坏等问题，提出CLAP框架，通过分割段落、解析共指链生成局部伪查询并融合全局与子主题信号，提升密集检索器性能，尤其在域外场景表现突出，nDCG@10最高提升20.68%。
摘要：
基于大语言模型（LLM）的段落扩展在增强第一阶段检索方面已显示出潜力，但由于语义漂移和与预训练语义空间的错位，其在密集检索器上的表现往往不佳。此外，段落中通常只有一部分与查询相关，其余部分会引入噪声——而破坏共指连续性的分块技术会加剧这一问题。我们提出了共指链接增强的段落检索（CLAP），这是一种轻量级的基于LLM的扩展框架，它将段落分割为连贯的块，解析共指链，并生成与密集检索器表示对齐的局部伪查询。全局主题信号与细粒度子主题信号的简单融合在跨领域实现了稳健的性能。即使检索器强度增加，CLAP也能持续带来增益，使密集检索器能够匹配或超越BM25 + MonoT5-3B等第二阶段排序器，nDCG@10绝对提升高达20.68%。这些改进在域外设置中尤为显著，而依赖领域知识的传统基于LLM的扩展方法在此类设置中往往表现不佳。CLAP采用以逻辑为中心的流水线，实现了稳健、领域无关的泛化。
英文摘要：
Large Language Model (LLM)-based passage expansion has shown promise for enhancing first-stage retrieval, but often underperforms with dense retrievers due to semantic drift and misalignment with their pretrained semantic space. Beyond this, only a portion of a passage is typically relevant to a query, while the rest introduces noise--an issue compounded by chunking techniques that break coreference continuity. We propose Coreference-Linked Augmentation for Passage Retrieval (CLAP), a lightweight LLM-based expansion framework that segments passages into coherent chunks, resolves coreference chains, and generates localized pseudo-queries aligned with dense retriever representations. A simple fusion of global topical signals and fine-grained subtopic signals achieves robust performance across domains. CLAP yields consistent gains even as retriever strength increases, enabling dense retrievers to match or surpass second-stage rankers such as BM25 + MonoT5-3B, with up to 20.68% absolute nDCG@10 improvement. These improvements are especially notable in out-of-domain settings, where conventional LLM-based expansion methods relying on domain knowledge often falter. CLAP instead adopts a logic-centric pipeline that enables robust, domain-agnostic generalization.
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。
