发表机构
University of Glasgow（格拉斯哥大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对传统检索器表层匹配导致的需求解读差距问题，提出生成式嵌入模型GEM，将推理与检索统一，在相关任务上性能优于基线，还可通过提示扩展测试时计算提升效果。
AI 中文摘要
现代大型语言模型（LLMs）在推理和指令遵循方面表现出色，使用户能够表达复杂多样的信息需求。然而，传统检索器主要依赖查询与文档之间的表层匹配，导致用户表达需求的方式与检索器解读需求的方式之间的差距不断扩大。在本文中，我们提出了GEM，一种生成式嵌入模型，它通过自身知识显式推理用户意图和相关性标准来增强检索。GEM在单一模型中统一了生成与嵌入：它首先对查询进行推理，然后添加一个嵌入标记以编码丰富的上下文用于检索。在推理密集型和指令遵循型检索任务上进行评估，GEM展示了其推理增强检索的有效性，优于其非推理变体以及使用大得多模型的基线。此外，GEM的生成特性允许通过提示进行测试时计算扩展，以进一步提升检索性能。我们的代码可在以下网址获取：this https URL。
英文摘要
Modern LLMs excel at reasoning and instruction following, enabling users to express complex and diverse information needs. However, conventional retrievers largely rely on surface-level matching between queries and documents, resulting in a growing gap between how users express their needs and how retrievers interpret them. In this paper, we present GEM, a generative embedding model that augments retrieval through its own knowledge by explicitly reasoning about user intent and relevance criteria. GEM unifies generation and embedding within a single model: it first reasons over the query, then appends an embedding token to encode the enriched context for retrieval. Evaluated on reasoning-intensive and instruction-following retrieval tasks, GEM demonstrates the effectiveness of its reasoning-augmented retrieval, outperforming its non-reasoning variant and matching baselines using substantially larger models. Furthermore, GEM's generative nature allows test-time compute scaling via prompting to further enhance retrieval performance. Our code is available at: https://anonymous.4open.science/r/GEM.
