Advancing Multi-Agent RAG Systems with Minimalist Reinforcement Learning

[2505.17086] Advancing Multi-Agent RAG Systems with Minimalist Reinforcement Learning Skip to main content Search Submit Donate Log in Search arXiv Press Enter to search · Advanced search Computer Science > Computation and Language arXiv:2505.17086 (cs) [Submitted on 20 May 2025 ( v1 ), last revised 14 Apr 2026 (this version, v4)] Title: Advancing Multi-Agent RAG Systems with Minimalist Reinforcement Learning Authors: Yihong Wu , Liheng Ma , Muzhi Li , Jiaming Zhou , Lei Ding , Jianye Hao , Ho-fung Leung , Irwin King , Yingxue Zhang , Jian-Yun Nie View a PDF of the paper titled Advancing Multi-Agent RAG Systems with Minimalist Reinforcement Learning, by Yihong Wu and 9 other authors View PDF HTML (experimental) Abstract: Large Language Models (LLMs) equipped with modern Retrieval-Augmented Generation (RAG) systems often employ multi-turn interaction pipelines to interface with search engines for complex reasoning tasks. However, such multi-turn interactions inevitably produce long intermediate contexts, as context length grows exponentially with exploration depth. This leads to a well-known limitation of LLMs: their difficulty in effectively leveraging information from long contexts. This problem is further amplified in RAG systems that depend on in-context learning, where few-shot demonstrations must also be included in the prompt, compounding the context-length bottleneck. To address these challenges, we propose Mujica-MyGo, a unified framework for efficient multi-turn reasoning in RAG. Inspired by the divide-and-conquer principle, we introduce Mujica (Multi-hop Joint Intelligence for Complex Question Answering), a multi-agent RAG workflow that decomposes multi-turn interactions into cooperative sub-interactions, thereby mitigating long-context issues. To eliminate the dependency on in-context learning, we further develop MyGO (Minimalist Policy Gradient Optimization), a lightweight and efficient reinforcement learning algorithm that enables effective post-training of LLMs within complex RAG pipelines. We provide theoretical guarantees for MyGO's convergence to the optimal policy. Empirical evaluations across diverse question-answering benchmarks, covering both text corpora and knowledge graphs, show that Mujica-MyGO achieves superior performance. Comments: AAMAS 2026 Subjects: Computation and Language (cs.CL) Cite as: arXiv:2505.17086 [cs.CL] (or arXiv:2505.17086v4 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2505.17086 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Yihong Wu [ view email ] [v1] Tue, 20 May 2025 18:33:03 UTC (429 KB) [v2] Sun, 13 Jul 2025 17:26:34 UTC (467 KB) [v3] Mon, 24 Nov 2025 00:33:29 UTC (503 KB) [v4] Tue, 14 Apr 2026 00:27:30 UTC (503 KB) Full-text links: Access Paper: View a PDF of the paper titled Advancing Multi-Agent RAG Systems with Minimalist Reinforcement Learning, by Yihong Wu and 9 other authors View PDF HTML (experimental) TeX Source view license Current browse context: cs.CL < prev | next > new | recent | 2025-05 Change to browse by: cs References & Citations NASA ADS Google Scholar Semantic Scholar export BibTeX citation Loading... BibTeX formatted citation × loading... Data provided by: Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer ( What is the Explorer? ) Connected Papers Toggle Connected Papers ( What is Connected Papers? ) Litmaps Toggle Litmaps ( What is Litmaps? ) scite.ai Toggle scite Smart Citations ( What are Smart Citations? ) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv ( What is alphaXiv? ) Links to Code Toggle CatalyzeX Code Finder for Papers ( What is CatalyzeX? ) DagsHub Toggle DagsHub ( What is DagsHub? ) GotitPub Toggle Gotit.pub ( What is GotitPub? ) Huggingface Toggle Hugging Face ( What is Huggingface? ) ScienceCast Toggle ScienceCast ( What is ScienceCast? ) Demos Demos Replicate Toggle Replicate ( What is Replicate? ) Spaces Toggle Hugging Face Spaces ( What is Spaces? ) Spaces Toggle TXYZ.AI ( What is TXYZ.AI? ) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower ( What are Influence Flowers? ) Core recommender toggle CORE Recommender ( What is CORE? ) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs . Which authors of this paper are endorsers? | Disable MathJax ( What is MathJax? ) We gratefully acknowledge support from our major funders , member institutions , , and all contributors. About · Help · Contact · Subscribe · Copyright · Privacy · Accessibility · Operational Status (opens in new tab) Major funding support from

Related documents

• Reinforcement learning control of quantum error correction.#395987
• RL2ML: Finite-Rollout Surrogate Objectives from Reinforcement Learning to Maximum Likelihood#238630
• Achieving Scale-Invariant Reinforcement Learning Performance with Reward Range Normalization#663276
• Achieving Scale-Invariant Reinforcement Learning Performance with Reward Range Normalization#660471
• A shortcut to an optimal quantum linear system solver#128293