mlsanityData to 14 Oct 2026

Multi-Agent Collaboration via Evolving Orchestration

arXiv 2505.19591 ↗First posted 26 May 2025cs.CL cs.AI cs.MAcited by 108

On the map: LLM agents and tool use → General LLM agent frameworks and benchmarks

Abstract (arXiv)

Large language models (LLMs) have achieved remarkable results across diverse downstream tasks, but their monolithic nature restricts scalability and efficiency in complex problem-solving. While recent research explores multi-agent collaboration among LLMs, most approaches rely on static organizational structures that struggle to adapt as task complexity and agent numbers grow, resulting in coordination overhead and inefficiencies. To this end, we propose a puppeteer-style paradigm for LLM-based multi-agent collaboration, where a centralized orchestrator ("puppeteer") dynamically directs agents ("puppets") in response to evolving task states. This orchestrator is trained via reinforcement learning to adaptively sequence and prioritize agents, enabling flexible and evolvable collective reasoning. Experiments on closed- and open-domain scenarios show that this method achieves superior performance with reduced computational costs. Analyses further reveal that the key improvements consistently stem from the emergence of more compact, cyclic reasoning structures under the orchestrator's evolution. Our code is available at https://github.com/OpenBMB/ChatDev/tree/puppeteer.

Key facts. Cited by 108 papers in the map (data through 14 Oct 2026). Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use. Builds on 72 of its 88 references that are in the map. First posted on arXiv 26 May 2025.

Builds on

72 of 88 references.

Closest ancestors · 13

Same direction, newest first

  1. 2025MAS-GPT: Training LLMs to Build LLM-based Multi-Agent Systems5 Mar
  2. 2024Optima: Optimizing Effectiveness and Efficiency for LLM-Based Multi-Agent System10 Oct
  3. 2024Automated Design of Agentic Systems15 Aug
  4. 2024Internet of Agents: Weaving a Web of Heterogeneous Agents for Collaborative Intelligence9 Jul
  5. 2024EvoAgent: Towards Automatic Multi-Agent Generation via Evolutionary Algorithms20 Jun
  6. 2024Scaling Large Language Model-based Multi-Agent Collaboration11 Jun
  7. 2024Agent Planning with World Knowledge Model23 May
  8. 2023Language Agent Tree Search Unifies Reasoning Acting and Planning in Language Models6 Oct
  9. 2023A Dynamic LLM-Powered Agent Network for Task-Oriented Agent Collaboration3 Oct
  10. 2023Agents: An Open-source Framework for Autonomous Language Agents14 Sep
  11. 2023AgentVerse: Facilitating Multi-Agent Collaboration and Exploring Emergent Behaviors21 Aug
  12. 2023MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework1 Aug
1 more
  1. 2023CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Society31 Mar

Other references · 59

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  3. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  4. 2020Scaling Laws for Neural Language Modelscited by 6,586
  5. 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,276
  6. 2023Toolformer: Language Models Can Teach Themselves to Use Toolscited by 4,182
  7. 2023Let's Verify Step by Stepcited by 3,789
  8. 2023Generative Agents: Interactive Simulacra of Human Behaviorcited by 3,568
  9. 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,442
  10. 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Modelscited by 3,211
  11. 2023Sparks of Artificial General Intelligence: Early experiments with GPT-4cited by 2,592
  12. 2022Emergent Abilities of Large Language Modelscited by 2,512
38 more
  1. 2022Do As I Can, Not As I Say: Grounding Language in Robotic Affordancescited by 2,348
  2. 2023Voyager: An Open-Ended Embodied Agent with Large Language Modelscited by 2,191
  3. 2023WebArena: A Realistic Web Environment for Building Autonomous Agentscited by 2,006
  4. 2023AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversationcited by 1,988
  5. 2023ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIscited by 1,947
  6. 2024MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmarkcited by 1,848
  7. 2023Improving Factuality and Reasoning in Language Models through Multiagent Debatecited by 1,784
  8. 2020ALFWorld: Aligning Text and Embodied Environments for Interactive Learningcited by 1,205
  9. 2023ChatDev: Communicative Agents for Software Developmentcited by 919
  10. 2023Graph of Thoughts: Solving Elaborate Problems with Large Language Modelscited by 886
  11. 2023Reasoning with Language Model is Planning with World Modelcited by 861
  12. 2019ALFRED: A Benchmark for Interpreting Grounded Instructions for Everyday Taskscited by 783
  13. 2022PAL: Program-aided Language Modelscited by 652
  14. 2025Why Do Multi-Agent LLM Systems Fail?cited by 604
  15. 2022LLM-Planner: Few-Shot Grounded Planning for Embodied Agents with Large Language Modelscited by 462
  16. 2024AFlow: Automating Agentic Workflow Generationcited by 432
  17. 2023Toxicity in ChatGPT: Analyzing Persona-assigned Language Modelscited by 360
  18. 2023MedAgents: Large Language Models as Collaborators for Zero-shot Medical Reasoningcited by 316
  19. 2023On the Planning Abilities of Large Language Models : A Critical Investigationcited by 312
  20. 2023Large Language Models as Tool Makerscited by 303
  21. 2023Role-Play with Large Language Modelscited by 294
  22. 2023ToolQA: A Dataset for LLM Question Answering with External Toolscited by 274
  23. 2023GPT4Tools: Teaching Large Language Model to Use Tools via Self-instructioncited by 248
  24. 2024Reinforcement learningcited by 247
  25. 2023PromptAgent: Strategic Planning with Language Models Enables Expert-level Prompt Optimizationcited by 233
  26. 2022Pre-Trained Language Models for Interactive Decision-Makingcited by 221
  27. 2024Agent Hospital: A Simulacrum of Hospital with Evolvable Medical Agentscited by 218
  28. 2023OpenAGI: When LLM Meets Domain Expertscited by 197
  29. 2024AlphaMath Almost Zero: Process Supervision without Processcited by 162
  30. 2023Beyond Chinchilla-Optimal: Accounting for Inference in Language Model Scaling Lawscited by 159
  31. 2023Language Agents with Reinforcement Learning for Strategic Play in the Werewolf Gamecited by 152
  32. 2023War and Peace (WarAgent): Large Language Model-based Multi-Agent Simulation of World Warscited by 145
  33. 2024HelpSteer2-Preference: Complementing Ratings with Preferencescited by 130
  34. 2020RODE: Learning Roles to Decompose Multi-Agent Taskscited by 125
  35. 2023ToolChain*: Efficient Action Space Navigation in Large Language Models with A* Searchcited by 105
  36. 2023On Generative Agents in Recommendationcited by 90
  37. 2023LASER: LLM Agent with State-Space Exploration for Web Navigationcited by 66
  38. 2023RET-LLM: Towards a General Read-Write Memory for Large Language Modelscited by 64

50 most cited of 59 All 59 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 21; 2026: 87.

108 papers.

Sort:

By area:

  1. 2025A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligencecited by 170
  2. 2025Towards a Science of Scaling Agent Systemscited by 138
  3. 2025A Survey of Context Engineering for Large Language Modelscited by 126
  4. 2025LEGOMem: Modular Procedural Memory for Multi-agent LLM Systems for Workflow Automationcited by 40
  5. 2026MAS-Orchestra: Understanding and Improving Multi-Agent Reasoning Through Holistic Orchestration and Controlled Benchmarkscited by 32
  6. 2025Learning to Orchestrate Agents in Natural Language with the Conductorcited by 23
  7. 2026The Orchestration of Multi-Agent Systems: Architectures, Protocols, and Enterprise Adoptioncited by 15
  8. 2026AgentConductor: Topology Evolution for Multi-Agent Competition-Level Code Generationcited by 13
  9. 2025Eigen-1: Adaptive Multi-Agent Refinement with Monitor-Based RAG for Scientific Reasoningcited by 11
  10. 2025Don't Throw Away Your Pretrained Modelcited by 10
  11. 2025Terrarium: Revisiting the Blackboard for Multi-Agent Safety, Privacy, and Security Studiescited by 10
  12. 2025HiMA-Ecom: Enabling Joint Training of Hierarchical Multi-Agent E-commerce Assistantscited by 9
38 more
  1. 2026FlowSteer: Towards Agents Designing Agentic Workflows via Reinforced Progressive Canvas Editingcited by 9
  2. 2026SkillGraph: Self-Evolving Multi-Agent Collaboration with Multimodal Graph Topologycited by 9
  3. 2025An Information Theoretic Perspective on Agentic System Designcited by 8
  4. 2026Agentic Environment Engineering for Large Language Models: A Survey of Environment Modeling, Synthesis, Evaluation, and Applicationcited by 8
  5. 2025Metacognitive Self-Correction for Multi-Agent System via Prototype-Guided Next-Execution Reconstructioncited by 7
  6. 2025The Era of Agentic Organization: Learning to Organize with Language Modelscited by 7
  7. 2025Agentic Services Computingcited by 6
  8. 2025Optimal-Agent-Selection: State-Aware Routing Framework for Efficient Multi-Agent Collaborationcited by 6
  9. 2025UFO3: Weaving the Digital Agent Galaxycited by 6
  10. 2026Uno-Orchestra: Parsimonious Agent Routing via Selective Delegationcited by 6
  11. 2026SkillFlow: Flow-Driven Recursive Skill Evolution for Agentic Orchestrationcited by 6
  12. 2026Reinforcement Learning for LLM-based Multi-Agent Systems through Orchestration Tracescited by 5
  13. 2026TacoMAS: Test-Time Co-Evolution of Topology and Capability in LLM-based Multi-Agent Systemscited by 5
  14. 2026Skill-MAS: Evolving Meta-Skill for Automatic Multi-Agent Systemscited by 5
  15. 2025InfiAgent: Self-Evolving Pyramid Agent Framework for Infinite Scenarioscited by 4
  16. 2026Don't Trust Stubborn Neighbors: A Security Framework for Agentic Networkscited by 4
  17. 2026MARCH: Multi-Agent Radiology Clinical Hierarchy for CT Report Generationcited by 4
  18. 2026Multi-Agent Transactive Memorycited by 4
  19. 2026Sakana Fugu Technical Reportcited by 4
  20. 2026MATA: A Trainable Hierarchical Automaton System for Multi-Agent Visual Reasoningcited by 3
  21. 2026Agent-Fence: Mapping Security Vulnerabilities Across Deep Research Agentscited by 3
  22. 2026Experience as a Compass: Multi-agent RAG with Evolving Orchestration and Agent Promptscited by 3
  23. 2026Federation over Text: Insight Sharing for Multi-Agent Reasoningcited by 3
  24. 2026Learning What to Retain: Gated-Memory Routing for Efficient Collaboration in Multi-Agent LLM Systemscited by 3
  25. 2025Orchestrator: Active Inference for Multi-Agent Systems in Long-Horizon Taskscited by 2
  26. 2026Beyond Rule-Based Workflows: An Information-Flow-Orchestrated Multi-Agents Paradigm via Agent-to-Agent Communication from CORALcited by 2
  27. 2026Enhancing Model Context Protocol (MCP) with Context-Aware Server Collaborationcited by 2
  28. 2026The Single-Multi Evolution Loop for Self-Improving Model Collaboration Systemscited by 2
  29. 2026Towards Adaptive, Scalable, and Robust Coordination of LLM Agents: A Dynamic Ad-Hoc Networking Perspectivecited by 2
  30. 2026MASRubric: Auditing Information Flow in Multi-Agent Systems with Failure-Distilled Pitfall Rubricscited by 2
  31. 2026HiMAP-Travel: Hierarchical Multi-Agent Planning for Long-Horizon Constrained Travelcited by 2
  32. 2026Small Model as Master Orchestrator: Learning Unified Agent-Tool Orchestration with Parallel Subtask Decompositioncited by 2
  33. 2026MetaAgent-X : Breaking the Ceiling of Automatic Multi-Agent Systems via End-to-End Reinforcement Learningcited by 2
  34. 2026Harnesses for Inference-Time Alignment over Execution Trajectoriescited by 2
  35. 2026Multi2: Hierarchical Multi-Agent Decision-Making with LLM-Based Agents in Interactive Environmentscited by 2
  36. 2026Cost-Utility Alignment in LLM Agent Trajectories:Profiling,Attribution,Diagnosis,Adaptation,and Evaluationcited by 2
  37. 2026Rethinking the Evaluation of Efficiency Methods for Multi-Agent Systemscited by 2
  38. 2026OmniHarness: Harnessing Generalizable Visual Generation via Symbolic Policy Learningcited by 2

50 most cited of 108 All 108 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026Adaptive Orchestration: Scalable Self-Evolving Multi-Agent Systems
  2. 2024MorphAgent: Empowering Agents through Self-Evolving Profiles and Decentralized Collaboration
  3. 2025AgentNet: Decentralized Evolutionary Coordination for LLM-based Multi-Agent Systems
  4. 2025Dynamic Strategy Adaptation in Multi-Agent Environments with Large Language Models
  5. 2025OPTAGENT: Optimizing Multi-Agent LLM Interactions Through Verbal Reinforcement Learning for Enhanced Reasoning
  6. 2026Multi-Agent Coordination Adaptation via Structure-Guided Orchestration
  7. 2026Orchestra-o1: Omnimodal Agent Orchestration
  8. 2025Reasoning-Aware Prompt Orchestration: A Foundation Model for Multi-Agent Language Model Coordination