Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning
This work addresses the challenge of enhancing LLM reasoning with real-time retrieval for tasks requiring external knowledge, representing an incremental advance in retrieval-augmented generation methods.
The paper tackles the problem of LLMs inefficiently using search engines during reasoning by introducing Search-R1, a reinforcement learning framework that trains LLMs to autonomously generate search queries during step-by-step reasoning, resulting in performance improvements of 41% and 20% over RAG baselines on question-answering datasets.
Efficiently acquiring external knowledge and up-to-date information is essential for effective reasoning and text generation in large language models (LLMs). Prompting advanced LLMs with reasoning capabilities to use search engines during inference is often suboptimal, as the LLM might not fully possess the capability on how to interact optimally with the search engine. This paper introduces Search-R1, an extension of reinforcement learning (RL) for reasoning frameworks where the LLM learns to autonomously generate (multiple) search queries during step-by-step reasoning with real-time retrieval. Search-R1 optimizes LLM reasoning trajectories with multi-turn search interactions, leveraging retrieved token masking for stable RL training and a simple outcome-based reward function. Experiments on seven question-answering datasets show that Search-R1 improves performance by 41% (Qwen2.5-7B) and 20% (Qwen2.5-3B) over various RAG baselines under the same setting. This paper further provides empirical insights into RL optimization methods, LLM choices, and response length dynamics in retrieval-augmented reasoning. The code and model checkpoints are available at https://github.com/PeterGriffinJin/Search-R1.