理解环境感知的信息检索行为
Understanding the Behaviors of Environment-aware Information Retrieval
AI总结：
通过强化学习使LLM适应不同检索器的查询策略，发现不同检索器偏好不同查询风格，并提出分支式滚动技术提升训练稳定性。
摘要：
最近的检索增强生成（RAG）方法在处理复杂查询方面展示了强大的能力，但当前研究忽略了一个关键挑战：不同的检索器需要根本不同的查询制定策略才能达到最佳性能。在这项工作中，我们首次系统分析了LLM如何通过强化学习（RL）学习适应不同检索器的查询制定策略。我们的实证研究表明，RL有效地教会了LLM根据特定检索器特征定制其查询。我们发现不同的检索器表现出令人惊讶的不同最优查询风格（例如，描述性 vs. 问题式），表明为一种检索器学习的策略对另一种检索器无效。我们进一步表明，通过结合检索器特定的人类指导和扩大模型规模可以提升性能。为了促进多检索步骤轨迹的学习，我们引入了一种基于分支的滚动技术，提高了训练稳定性。我们的工作为构建真正检索器感知的RAG系统提供了首个实证证据和可操作的见解。代码和资源可在 https://github.com/LCO-Embedding/Envs-aware-Information-Retrieval 获取。
英文摘要：
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.
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。
