mlsanityData to 13 Oct 2026

Closed-Loop Vision-Language Planning for Multi-Agent Coordination

arXiv 2502.10148 ↗First posted 14 Feb 2025cs.AI cs.MAcited by 2

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

Abstract (arXiv)

Cooperative multi-agent reinforcement learning (MARL) struggles with sample efficiency, interpretability, and generalization. While Large Language Models (LLMs) offer powerful planning capabilities, their application has been hampered by a reliance on text-only inputs and a failure to handle the non-Markovian, partially observable nature of multi-agent tasks. We introduce COMPASS, a multi-agent framework that overcomes these limitations by integrating Vision-Language Models (VLMs) for decentralized, closed-loop decision-making. COMPASS dynamically generates and refines interpretable, code-based strategies stored in a skill library that is bootstrapped from expert demonstrations. To ensure robust coordination, it propagates entity information through a structured multi-hop communication protocol, allowing teams to build a coherent understanding from partial observations. Evaluated on the challenging SMACv2 benchmark, COMPASS significantly outperforms state-of-the-art MARL baselines. Notably, in the symmetric Protoss 5v5 task, COMPASS achieved a 57% win rate, a 30 percentage point advantage over QMIX (27%). Project page can be found at https://stellar-entremet-1720bb.netlify.app/.

Key facts. Cited by 2 papers in the map (data through 13 Oct 2026). Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use. Builds on 26 of its 55 references that are in the map. First posted on arXiv 14 Feb 2025.

Builds on

26 of 55 references.

References · 26

  1. 2022Large Language Models are Zero-Shot Reasonerscited by 4,348
  2. 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,156
  3. 2017Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environmentscited by 1,525
  4. 2022Code as Policies: Language Model Programs for Embodied Controlcited by 1,186
  5. 2024Executable Code Actions Elicit Better LLM Agentscited by 548
  6. 2020Monotonic Value Function Factorisation for Deep Multi-Agent Reinforcement Learningcited by 478
  7. 2018VirtualHome: Simulating Household Activities via Programscited by 414
  8. 2019QTRAN: Learning to Factorize with Transformation for Cooperative Multi-Agent Reinforcement Learningcited by 374
  9. 2019On the Utility of Learning about Humans for Human-AI Coordinationcited by 346
  10. 2017StarCraft II: A New Challenge for Reinforcement Learningcited by 269
  11. 2020QPLEX: Duplex Dueling Multi-Agent Q-Learningcited by 268
  12. 2019Google Research Football: A Novel Reinforcement Learning Environmentcited by 258
14 more
  1. 2023Large Language Models as Commonsense Knowledge for Large-Scale Task Planningcited by 231
  2. 2018Fully Decentralized Multi-Agent Reinforcement Learning with Networked Agentscited by 190
  3. 2019MAVEN: Multi-Agent Variational Explorationcited by 183
  4. 2023RoCo: Dialectic Multi-Robot Collaboration with Large Language Modelscited by 171
  5. 2022Multi-Agent Reinforcement Learning is a Sequence Modeling Problemcited by 154
  6. 2023Large Language Models Play StarCraft II: Benchmarks and A Chain of Summarization Approachcited by 79
  7. 2024Learning Multi-Agent Communication from Graph Modeling Perspectivecited by 48
  8. 2024CooHOI: Learning Cooperative Human-Object Interaction with Manipulated Object Dynamicscited by 43
  9. 2020Randomized Entity-wise Factorization for Multi-Agent Reinforcement Learningcited by 38
  10. 2023Learning Multi-Agent Communication with Contrastive Learningcited by 9
  11. 2024Boosting Sample Efficiency and Generalization in Multi-agent Reinforcement Learning via Equivariancecited by 9
  12. 2024Learning Multi-Agent Loco-Manipulation for Long-Horizon Quadrupedal Pushingcited by 9
  13. 2025WFCRL: A Multi-Agent Reinforcement Learning Benchmark for Wind Farm Controlcited by 7
  14. 2024AgentMixer: Multi-Agent Correlated Policy Factorizationcited by 5

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026DiLLS: Interactive Diagnosis of LLM-based Multi-agent Systems via Layered Summary of Agent Behaviorscited by 7
  2. 2026Beyond Individual Intelligence: Surveying Collaboration, Failure Attribution, and Self-Evolution in LLM-based Multi-Agent Systemscited by 5

Similar, not cited

Close in wording, no citation either way

  1. 2026Systematic Multi-Agent Vision-and-Language Navigation: Formulation, Benchmark, and Method
  2. 2026Controlling Collectives of AI Agents in Reasoning Space with Spatial Transformers
  3. 2024Towards Collaborative Intelligence: Propagating Intentions and Reasoning for Multi-Agent Coordination with Large Language Models
  4. 2025Benchmarking LLMs' Swarm intelligence
  5. 2025Decentralized Multi-Agent Goal Assignment for Path Planning using Large Language Models
  6. 2026Cooperative Multi-Agent Vision-Language-Action Models via Reinforced Fine Tuning
  7. 2025VAGEN: Reinforcing World Model Reasoning for Multi-Turn VLM Agents
  8. 2026Benchmarking Open-Ended Multi-Agent Coordination in Language Agents