让检索模型学会推理隐含关联，提升复杂查询的准确率
Imagine All The Relevance: Scenario-Profiled Indexing with Knowledge Expansion for Dense Retrieval
- 将文档拆解为场景单元，显式建模推理路径
- 用大模型生成场景增强数据，蒸馏出高效场景生成器
- 支持复杂查询与RAG应用，适合需要深度理解的场景
现有密集检索模型在需要推理的任务中表现不佳，因其难以捕捉超越表面语义的隐含相关性。为此，我们提出情景-画像索引与知识扩展框架（SPIKE），通过将文档分解为基于情景的检索单元，显式索引隐含相关性。SPIKE将文档组织为包含推理过程的情景，以揭示假设信息需求与文档内容间的深层联系。利用强大教师大模型构建情景增强数据集，并将推理能力蒸馏至小型高效的情景生成器。推理时，SPIKE同时融合情景级与文档级相关性，实现具备推理意识的检索。大量实验表明，SPIKE在多种查询类型和密集检索器上均持续提升性能，还通过提供情景信息改善用户检索体验，并为检索增强生成（RAG）中的大模型提供有价值上下文。
原文摘要 · Abstract (English)
Existing dense retrieval models struggle with reasoning-intensive retrieval task as they fail to capture implicit relevance that requires reasoning beyond surface-level semantic information. To address these challenges, we propose Scenario-Profiled Indexing with Knowledge Expansion (SPIKE), a dense retrieval framework that explicitly indexes implicit relevance by decomposing documents into scenario-based retrieval units. SPIKE organizes documents into scenario, which encapsulates the reasoning process necessary to uncover implicit relationships between hypothetical information needs and document content. SPIKE constructs a scenario-augmented dataset using a powerful teacher large language model (LLM), then distills these reasoning capabilities into a smaller, efficient scenario generator. During inference, SPIKE incorporates scenario-level relevance alongside document-level relevance, enabling reasoning-aware retrieval. Extensive experiments demonstrate that SPIKE consistently enhances retrieval performance across various query types and dense retrievers. It also enhances the retrieval experience for users through scenario and offers valuable contextual information for LLMs in retrieval-augmented generation (RAG).
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。