Multi-Agent Reasoning

Follow this topic Ask about this topic

Multi-agent reasoning examines how AI agents combine knowledge and perspectives to solve problems that may exceed one agent’s abilities, including mathematical and scientific question answering, games, robotics, and clinical decisions. Common approaches include role-specialized agents, chain-of-thought debate and critique, voting or mixture-of-agents aggregation, and learned routing that decides who should contribute and when. Other work trains collaboration with reinforcement learning or distills group reasoning into a single model, while benchmarks test coordination under partial information and strategic interaction. A central tension is that collaboration can correct individual errors, but communication costs, correlated mistakes, and pressure to conform can erase its gains; when it helps depends on the task, evidence quality, and compute budget.

Latest papers

233 papers29 papers in the last four weeks, up 142% on the four weeks before. 0.2% of all new papers.

Unifying Temporal and Structural Credit Assignment in LLM-Based Multi-Agent Prompt Optimization
May 28, 2026Wenwu Li, Yuran Song, Mingze Zhao +2Multi-Agent LLM Systems Multi-Agent System Optimization
Tongji University, Shanghai, China · The University of Hong Kong, Hong Kong, China
MINDGAMES: A Live Arena for Evaluating Social and Strategic Reasoning in Multi-Agent LLMs
May 28, 2026Kevin Wang, Anna Thöni, Benjamin Kempinski +50LLM Agent Evaluation Strategic Reasoning Risks in Language Models
DynaGraph: Lightweight Multi-Model Interaction Framework via Dynamic Topological Reconfiguration
May 28, 2026Yanxing Guo, Zihao Zheng, Fangzhou Wu +4Multi-Agent LLM Systems LLM Routing
Peking University · Nanjing University · Beijing Advanced Innovation Center for Integrated Circuits +2
Beyond Consensus: Trace-Level Synthesis in Mixture of Agents
May 27, 2026Shreyas Fadnavis, Praitayini Kanakaraj, Felix WyssLanguage Model Ensembles Multi-Agent Reasoning
Bioscope AI
TRACER: Turn-level Regret Matching with Inner Reinforcement Credit for Cooperative Multi-LLM Reasoning
May 27, 2026Chusen Li, Zhou Liu, Shuigeng Zhou +1Regret Minimization Multi-Agent Reasoning
1Fudan University · 2Zhongguancun Academy · 3Academy for Advanced Interdisciplinary Studies, Peking University
Roles with Rails: Contract-Preserving Role Evolution in Multi-Agent Structured Reasoning
May 27, 2026Ling-Yue Ge, Lan-Zhe GuoMulti-Agent LLM Systems LLM Agent Orchestration
1National Key Laboratory for Novel Software Technology, Nanjing University, Nanjing, China · School of Intelligence Science and Technology, Nanjing University, Suzhou, China
MACReD: A Multi-Agent Collaborative Reasoning Framework for Reaction Diagram Parsing
May 27, 2026Chuang Tang, Chenhao Lin, Yin Xu +5VLM Reasoning Multi-Agent Reasoning
University of Science and Technology of China Hefei, China
PEAR: Permutation-Equivariant Adaptive Routing Multi-Agent Debate
May 26, 2026Yang Feng, Ziwei Xu, Xia Hu +1Multi-Agent Debate Multi-Agent Systems
University of Edinburgh · Shanghai AI Laboratory
Multi-Agent Systems are Mixtures of Experts: Who Becomes an Influencer?
May 25, 2026Franka Bause, Jonas Niederle, Martin Pawelczyk +1Multi-Agent LLM Systems Opinion Dynamics
CISPA Helmholtz Center for Information Security, Saarbrücken, Germany · Faculty of Computer Science, University of Vienna, Vienna, Austria.
DarkForest: Less Talk, Higher Accuracy for Multi-Agent LLMs
May 24, 2026Yi Li, Songtao Wei, Dongming Jiang +3Multi-Agent LLM Systems Multi-Agent System Optimization
University of Texas at Dallas · Independent Researcher · University of California, Davis
AgentFugue: Agent Scaling for Long-Horizon Tasks through Collective Reasoning
May 23, 2026Yuyang Hu, Hongjin Qian, Shuting Wang +5Multi-Agent Reasoning Long-Horizon LLM Agents
GSAI, Renmin University of China · Beijing Academy of Artificial Intelligence
Higher Order Reasoning for Collaborative Communicationless Mobile Robot Operations
May 21, 2026Jonathan Reasoner, Nicola BezzoBelief-Space Planning Multi-Robot Motion Planning
Department of Electrical and Computer Engineering, University of Virginia, Charlottesville, VA 22904, USA
What Do Agents Communicate? Characterizing Information Exchange in Multi-Agent Systems
May 19, 2026Yong Jin Chun, Iftekhar AhmedMulti-Agent Collaboration Multi-Agent Systems
Department of Informatics University of California, Irvine Irvine, CA, 92618
AutoResearchClaw: Self-Reinforcing Autonomous Research with Human-AI Collaboration
May 19, 2026Jiaqi Liu, Shi Qiu, Mairui Li +33Human-in-the-Loop AI Autonomous Scientific Discovery
UNC · Rutgers University · UC Santa Cruz +8
Conflict-Aware Multi-Agent Reasoning via Signed Graph Modeling
May 19, 2026Longgang He, Longzhu He, Shihao Peng +2Multi-Agent LLM Systems Multi-Agent Systems
Harbin Institute of Technology (Shenzhen) · Beijing University of Posts and Telecommunications
STAR-PólyaMath: Multi-Agent Reasoning under Persistent Meta-Strategic Supervision
May 19, 2026Jiaao Wu, Xian Zhang, Hanzhang Liu +3Multi-Agent Orchestration Multi-Agent Reasoning
1Tsinghua University · 2Microsoft Research · 3New York University +1
AMATA: Adaptive Multi-Agent Trajectory Alignment for Knowledge-Intensive Question Answering
May 17, 2026Taolin Zhang, Dongyang Li, Chen Chen +5Direct Preference Optimization Agentic RAG
School of Computer Science and Information Engineering, Hefei University of Technology · Shanghai University of Electric Power · Guangdong University of Finance and Economics +2
Towards In-Depth Root Cause Localization for Microservices with Multi-Agent Recursion-of-Thought
May 14, 2026Lingzhe Zhang, Tong Jia, Kangjin Wang +8Software Engineering Multi-Agent LLM Systems
Peking University, Beijing, China · Huawei Theory Lab, China
Hierarchical Attacks for Multi-Modal Multi-Agent Reasoning
May 13, 2026Hao Zhou, Tiru Wu, Yan Jiang +3Multimodal Robustness Adversarial Attacks
Predictive Maps of Multi-Agent Reasoning: A Successor-Representation Spectrum for LLM Communication Topologies
May 12, 2026Ethan Parks, Dalal AlharthiMulti-Agent Systems Multi-Agent Communication Topology
University of Arizona
The Bystander Effect in Multi-Agent Reasoning: Quantifying Cognitive Loafing in Collaborative Interactions
May 11, 2026Dahlia Shehata, Ming LiLLM Sycophancy Multi-Agent Collaboration
University of Waterloo Canada
UTS at PsyDefDetect: Multi-Agent Councils and Absence-Based Reasoning for Defense Mechanism Classification
May 10, 2026Dima Galat, Marian-Andrei RizoiuMulti-Agent LLM Systems Mental Health
University of Technology Sydney
MOTOR-Bench: A Real-world Dataset and Multi-agent Framework for Zero-shot Human Mental State Understanding
May 10, 2026Xiaoyu Yuan, Niklas Heikkala, Tiina Törmänen +3Theory of Mind AI Agent Benchmarks
University of Oulu
Robust Multi-Agent LLMs under Byzantine Faults
May 9, 2026Haejoon Lee, Vincent-Daniel Yun, Dimitra Panagou +1Multi-Agent LLM Systems Multi-Agent Coordination
Department of Robotics, University of Michigan, Ann Arbor, USA · Thomas Lord Department of Computer Science, University of Southern California, USA
AgentPSO: Evolving Agent Reasoning Skill via Multi-agent Particle Swarm Optimization
May 9, 2026Hyunmin Hwang, Jaemin Kim, Choonghan Kim +2LLM Agent Skill Learning Multi-Agent Reasoning
Graduate School of Artificial Intelligence, Korea Advanced Institute of Science and Technology, Seoul, South Korea.
Iterative Critique-and-Routing Controller for Multi-Agent Systems with Heterogeneous LLMs
May 9, 2026Wenzhi Fang, Liangqi Yuan, Guangchen Lan +2Multi-Agent LLM Systems Multi-Agent Coordination
Purdue University · Yonsei University
The Memory Curse: How Expanded Recall Erodes Cooperative Intent in LLM Agents
May 8, 2026Jiayuan Liu, Tianqin Li, Shiyi Du +7Social Dilemmas LLM Agent Memory
Carnegie Mellon University · Foundations of Cooperative AI Lab (FOCAL) · University of Michigan +1
MAVEN: Multi-Agent Verification-Elaboration Network with In-Step Epistemic Auditing
May 8, 2026Yinsheng Yao, Jiehao Tang, Zhaozhen Yang +1LLM Auditing Multi-Agent Debate
Tongji University Shanghai, China
SOM: Structured Opponent Modeling for LLM-based Agents via Structural Causal Model
May 8, 2026Shiyue Cao, Pei Xu, Likun Yang +3Opponent Modeling Multi-Agent Reasoning
School of Artificial Intelligence, University of Chinese Academy of Sciences & Institute of Automation, Chinese Academy of Sciences Beijing, China · National Key Laboratory of Cognition and Decision Intelligence for Complex Systems, Institute of Automation, Chinese Academy of Sciences Beijing, China · School of Artificial Intelligence, University of Chinese Academy of Sciences Beijing, China +1
Strat-Reasoner: Reinforcing Strategic Reasoning of LLMs in Multi-Agent Games
May 6, 2026Yidong He, Yutao Lai, Pengxu Yang +4Game-Playing Agents RL for Language Model Reasoning
School of Software Engineering, South China University of Technology · Department of Computer Science, University of Oxford

May 28, 2026J/K move · Enter open · S save

Unifying Temporal and Structural Credit Assignment in LLM-Based Multi-Agent Prompt Optimization

Wenwu Li, Yuran Song, Mingze Zhao, Bo Jin+1

Tongji University, Shanghai, China · The University of Hong Kong, Hong Kong, China

Multi-Agent LLM Systems Multi-Agent System Optimization Large Language Model-Guided Optimization Multi-Agent Systems Multi-Agent Reasoning

Open paper AskPDF arXiv

While Multi-Agent Systems (MAS) empower Large Language Models to tackle complex reasoning tasks through collaborative interaction, optimizing their dynamics remains a formidable challenge due to the discrete, non-differentiable nature of the computation graph and the sparsity of global supervisory signals. Existing black-box optimizers struggle to attribute trajectory-level failure to specific local components, resulting in inefficient, high-variance exploration. We argue that tractable MAS optimization needs structural inductive biases to disentangle error signals. We propose temporal and structural credit assignment, which decomposes the objective along two axes: (i) temporal credit, using state-space bottlenecks to identify critical rounds, and (ii) structural credit, using stationary role policies to isolate agent contributions. Leveraging these decomposed signals, we introduce a discrete, verbalized block coordinate descent algorithm for iterative refinement. Rather than indiscriminate global updates, it alternates between optimizing role prompts and aggregation protocols, using LLM-generated "proxy gradients" to target only the identified weak links. Across diverse reasoning benchmarks, our approach substantially reduces query complexity while improving performance, providing a principled and interpretable path toward self-improving MAS.