理解环境感知信息检索的行为

摘要

近期，检索增强生成（RAG）方法在处理复杂查询方面展现出强大能力，然而现有研究忽视了一个关键挑战：不同检索器需要截然不同的查询构建策略才能实现最优性能。在本研究中，我们首次系统分析了如何通过强化学习（RL）使大语言模型（LLM）学会为不同检索器自适应调整查询构建策略。我们的实证研究表明，RL能有效引导LLM针对特定检索器的特性定制查询。我们发现，不同检索器对最优查询风格（如描述型与疑问型）存在显著偏好差异，这意味着为某类检索器习得的策略对另一类检索器可能无效。进一步研究表明，融入检索器特定的人类指导以及扩大模型规模均可提升性能。为优化多步检索轨迹的学习过程，我们引入了一种基于分支的展开技术，有效提升了训练稳定性。本研究为构建真正具备检索器感知能力的RAG系统提供了首个实证依据与可操作见解。代码与资源详见 https://github.com/LCO-Embedding/Envs-aware-Information-Retrieval。

English

Recent retrieval-augmented generation (RAG) approaches have demonstrated strong capability in handling complex queries, yet current research overlooks a critical challenge: different retrievers require fundamentally different query formulation strategies for optimal performance. In this work, we present the first systematic analysis of how LLMs can learn to adapt their query formulation strategies for different retrievers via reinforcement learning (RL). Our empirical study reveals that RL effectively teaches an LLM to tailor its queries to specific retriever characteristics. We discover that different retrievers exhibit surprisingly distinct optimal query styles (e.g., descriptive vs. question-like), suggesting strategies learned for one retriever ineffective for another. We further show that performance can be enhanced by incorporating retriever-specific human guidance and by scaling model size. To facilitate learning over multi-retrieval-step trajectories, we introduce a branching-based rollout technique that improves training stability. Our work provides the first empirical evidence and actionable insights for building truly retriever-aware RAG systems. Code and resources are available at https://github.com/LCO-Embedding/Envs-aware-Information-Retrieval.