Evolve as a Team: Collaborative Self-Evolution for LLM-based Multi-Agent Systems

arXiv:2605.29790 · cs.MA, cs.AI · Submitted 2026-05-28 · Read on arXiv

cs.MA, cs.AI

Submitted: 2026-05-28

Updated: 2026-10-13

Code: https://github.com/zz-haooo/Meta-Team

Terminology

• Multi-Agent System — 2× in this paper · explained in 6 episodes
• Benchmark — 1× in this paper · explained in 2 episodes

Sources

• GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning
• Program Synthesis with Large Language Models
• BeyondSWE: Can Current Code Agent Survive Beyond Single-Repo Bug Fixing?
• GoAgent: Group-of-Agents Communication Topology Generation for LLM-based Multi-Agent Systems
• Evaluating Large Language Models Trained on Code
• Seeing the Whole Elephant: A Benchmark for Failure Attribution in LLM-based Multi-Agent Systems
• Scoring, Reasoning, and Selecting the Best! Ensembling Large Language Models via a Peer-Review Process
• Training Verifiers to Solve Math Word Problems
• SWE-Bench Pro: Can AI Agents Solve Long-Horizon Software Engineering Tasks?
• Magentic-One: A Generalist Multi-Agent System for Solving Complex Tasks
• FlowReasoner: Reinforcing Query-Level Meta-Agents
• ReCreate: Reasoning and Creating Domain Agents Driven by Experience
• RULER: What's the Real Context Size of Your Long-Context Language Models?
• EvoMAS: Evolutionary Generation of Multi-Agent Systems
• Rethinking Failure Attribution in Multi-Agent Systems: A Multi-Perspective Benchmark and Evaluation
• Aegis: Automated Error Generation and Attribution for Multi-Agent Systems
• Workflow-R1: Group Sub-sequence Policy Optimization for Multi-turn Workflow Construction
• Agentic Aggregation for Parallel Scaling of Long-Horizon Agentic Tasks
• Towards Self-Improving Error Diagnosis in Multi-Agent Systems
• AgencyBench: Benchmarking the Frontiers of Autonomous Agents in 1M-Token Real-World Contexts

Related papers

• Highway Congestion Reduction through Reinforcement Learning Based Eulerian Headway Control
• ORACLE-SWE: Quantifying the Contribution of Critical Information Signals on SWE Agents
• You Only Align Once: Propagating Cooperative Behaviors in Multi-Agent Systems through Seed Agents
• Deny Without Disabling: Authorization-Paired Evaluation and Control for Multi-Agent Systems
• MA-SAPO: Multi-Agent Reasoning for Score-Aware Prompt Optimization
• PeroMAS: A Multi-agent System of Perovskite Material Discovery