{"work":{"id":"460d8021-a193-45c3-89a7-b72f6767c87f","openalex_id":null,"doi":null,"arxiv_id":"2504.03160","raw_key":null,"title":"DeepResearcher: Scaling Deep Research via Reinforcement Learning in Real-world Environments","authors":null,"authors_text":"Yuxiang Zheng, Dayuan Fu, Xiangkun Hu, Xiaojie Cai, Lyumanshan Ye, Pengrui Lu","year":2025,"venue":"cs.AI","abstract":"Large Language Models (LLMs) equipped with web search capabilities have demonstrated impressive potential for deep research tasks. However, current approaches predominantly rely on either manually engineered prompts (prompt engineering-based) with brittle performance or reinforcement learning within controlled Retrieval-Augmented Generation (RAG) environments (RAG-based) that fail to capture the complexities of real-world interaction. In this paper, we introduce DeepResearcher, the first comprehensive framework for end-to-end training of LLM-based deep research agents through scaling reinforcement learning (RL) in real-world environments with authentic web search interactions. Unlike RAG-based approaches that assume all necessary information exists within a fixed corpus, our method trains agents to navigate the noisy, unstructured, and dynamic nature of the open web. We implement a specialized multi-agent architecture where browsing agents extract relevant information from various webpage structures and overcoming significant technical challenges. Extensive experiments on open-domain research tasks demonstrate that DeepResearcher achieves substantial improvements of up to 28.9 points over prompt engineering-based baselines and up to 7.2 points over RAG-based RL agents. Our qualitative analysis reveals emergent cognitive behaviors from end-to-end RL training, including the ability to formulate plans, cross-validate information from multiple sources, engage in self-reflection to redirect research, and maintain honesty when unable to find definitive answers. Our results highlight that end-to-end training in real-world web environments is not merely an implementation detail but a fundamental requirement for developing robust research capabilities aligned with real-world applications. We release DeepResearcher at https://github.com/GAIR-NLP/DeepResearcher.","external_url":"https://arxiv.org/abs/2504.03160","cited_by_count":null,"metadata_source":"pith","metadata_fetched_at":"2026-07-03T04:37:37.738752+00:00","pith_arxiv_id":"2504.03160","created_at":"2026-05-10T06:51:45.769659+00:00","updated_at":"2026-07-03T04:37:37.738752+00:00","title_quality_ok":true,"display_title":"DeepResearcher: Scaling Deep Research via Reinforcement Learning in Real-world Environments","render_title":"DeepResearcher: Scaling Deep Research via Reinforcement Learning in Real-world Environments"},"hub":{"state":{"work_id":"460d8021-a193-45c3-89a7-b72f6767c87f","tier":"hub","tier_reason":"10+ Pith inbound or 1,000+ external citations","pith_inbound_count":33,"external_cited_by_count":null,"distinct_field_count":7,"first_pith_cited_at":"2025-04-16T21:45:32+00:00","last_pith_cited_at":"2026-06-30T13:29:58+00:00","author_build_status":"not_needed","summary_status":"needed","contexts_status":"needed","graph_status":"needed","ask_index_status":"not_needed","reader_status":"not_needed","recognition_status":"not_needed","updated_at":"2026-08-22T02:49:29.487504+00:00","tier_text":"hub"},"tier":"hub","role_counts":[{"context_role":"background","n":3},{"context_role":"method","n":1}],"polarity_counts":[{"context_polarity":"background","n":3},{"context_polarity":"use_method","n":1}],"runs":{},"summary":{},"graph":{},"authors":[]}}