Understanding the Behaviors of Environment-aware Information Retrieval

ACL ARR 2026 January Submission7565 Authors

Keywords: information retrieval, query rewrite, reinforcement learning

Abstract: 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.

Paper Type: Long

Research Area: Information Extraction and Retrieval

Research Area Keywords: information retrieval, query rewrite, reinforcement learning

Contribution Types: Model analysis & interpretability, NLP engineering experiment

Languages Studied: English

Submission Number: 7565