Do LLM-derived graph priors improve multi-agent coordination?

• URL: http://arxiv.org/abs/2604.17191v1
• Date: Sun, 19 Apr 2026 01:40:39 GMT
• Title: Do LLM-derived graph priors improve multi-agent coordination?
• Abstract summary: Large language models (LLMs) can generate useful coordination graph priors for multi-agent reinforcement learning (MARL)<n>Our results provide the first quantitative evidence that LLM-derived graph priors can enhance coordination and adaptability in dynamic multi-agent environments.
• Score: 5.3569479406244325
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Multi-agent reinforcement learning (MARL) is crucial for AI systems that operate collaboratively in distributed and adversarial settings, particularly in multi-domain operations (MDO). A central challenge in cooperative MARL is determining how agents should coordinate: existing approaches must either hand-specify graph topology, rely on proximity-based heuristics, or learn structure entirely from environment interaction; all of which are brittle, semantically uninformed, or data-intensive. We investigate whether large language models (LLMs) can generate useful coordination graph priors for MARL by using minimal natural language descriptions of agent observations to infer latent coordination patterns. These priors are integrated into MARL algorithms via graph convolutional layers within a graph neural network (GNN)-based pipeline, and evaluated on four cooperative scenarios from the Multi-Agent Particle Environment (MPE) benchmark against baselines spanning the full spectrum of coordination modeling, from independent learners to state-of-the-art graph-based methods. We further ablate across five compact open-source LLMs to assess the sensitivity of prior quality to model choice. Our results provide the first quantitative evidence that LLM-derived graph priors can enhance coordination and adaptability in dynamic multi-agent environments, and demonstrate that models as small as 1.5B parameters are sufficient for effective prior generation.

Related papers

• Efficient Recommendations via Graph Coarsening and Label Propagation [50.71909189967895]
  Graph-based recommendations are widely adopted in real-world industrial applications.<n>One promising approach is graph coarsening, an adaptive graph reduction technique.<n>We propose a flexible two-stage diffusion framework that combines graph coarsening with multi-step label propagation.
  arXiv  Detail & Related papers  (2026-07-24T13:25:26Z)
• Aggregation-aware MLP: An Unsupervised Approach for Graph Message-passing [10.93155007218297]
  "AMLP" is an unsupervised framework that shifts the paradigm from directly crafting aggregation functions to making adaptive aggregation.<n>Our approach consists of two key steps: First, we utilize a graph reconstruction that facilitates high-order grouping effects, and second, we employ a single-layer network to encode varying degrees of heterophily.
  arXiv  Detail & Related papers  (2025-07-27T04:52:55Z)
• MultiAgentBench: Evaluating the Collaboration and Competition of LLM agents [59.825725526176655]
  Large Language Models (LLMs) have shown remarkable capabilities as autonomous agents.<n>Existing benchmarks either focus on single-agent tasks or are confined to narrow domains, failing to capture the dynamics of multi-agent coordination and competition.<n>We introduce MultiAgentBench, a benchmark designed to evaluate LLM-based multi-agent systems across diverse, interactive scenarios.
  arXiv  Detail & Related papers  (2025-03-03T05:18:50Z)
• Deep Meta Coordination Graphs for Multi-agent Reinforcement Learning [2.650735171795961]
  Deep meta coordination graphs (DMCG) for learning cooperative policies in multi-agent reinforcement learning (MARL)<n>DMCG captures useful higher-order and indirect relationships among agents.<n>It then employs a graph convolutional network module to learn powerful representations in an end-to-end manner.
  arXiv  Detail & Related papers  (2025-02-06T12:35:52Z)
• Graph Learning in the Era of LLMs: A Survey from the Perspective of Data, Models, and Tasks [25.720233631885726]
  integration of Graph Neural Networks (GNNs) and Large Language Models (LLMs) has emerged as a promising technological paradigm.<n>We leverage graph description texts with rich semantic context to fundamentally enhance Data quality.<n>This work serves as a foundational reference for researchers and practitioners looking to advance graph learning methodologies.
  arXiv  Detail & Related papers  (2024-12-17T01:41:17Z)
• Self-Clustering Hierarchical Multi-Agent Reinforcement Learning with Extensible Cooperation Graph [9.303181273699417]
  This paper proposes a novel hierarchical MARL model called Hierarchical Cooperation Graph Learning (HCGL) HCGL has three components: a dynamic Cooperation Graph (ECG) for achieving self-clustering cooperation; a group of graph operators for adjusting the topology of ECG; and an MARL for training these graph operators. In our experiments, the HCGL model has shown outstanding performance in multi-agent benchmarks with sparse rewards.
  arXiv  Detail & Related papers  (2024-03-26T19:19:16Z)
• Structured Cooperative Learning with Graphical Model Priors [98.53322192624594]
  We study how to train personalized models for different tasks on decentralized devices with limited local data. We propose "Structured Cooperative Learning (SCooL)", in which a cooperation graph across devices is generated by a graphical model. We evaluate SCooL and compare it with existing decentralized learning methods on an extensive set of benchmarks.
  arXiv  Detail & Related papers  (2023-06-16T02:41:31Z)
• Distributed Multi-Agent Reinforcement Learning Based on Graph-Induced Local Value Functions [7.6860514640178]
  We propose a general computationally efficient distributed framework for cooperative multi-agent reinforcement learning (MARL) We introduce three coupling graphs describing three types of inter-agent couplings in MARL. We propose two distributed RL approaches based on local value-functions derived from the coupling graphs.
  arXiv  Detail & Related papers  (2022-02-26T03:01:51Z)
• Soft Hierarchical Graph Recurrent Networks for Many-Agent Partially Observable Environments [9.067091068256747]
  We propose a novel network structure called hierarchical graph recurrent network(HGRN) for multi-agent cooperation under partial observability. Based on the above technologies, we proposed a value-based MADRL algorithm called Soft-HGRN and its actor-critic variant named SAC-HRGN.
  arXiv  Detail & Related papers  (2021-09-05T09:51:25Z)
• Efficient Model-Based Multi-Agent Mean-Field Reinforcement Learning [89.31889875864599]
  We propose an efficient model-based reinforcement learning algorithm for learning in multi-agent systems. Our main theoretical contributions are the first general regret bounds for model-based reinforcement learning for MFC. We provide a practical parametrization of the core optimization problem.
  arXiv  Detail & Related papers  (2021-07-08T18:01:02Z)
• Cooperative Policy Learning with Pre-trained Heterogeneous Observation Representations [51.8796674904734]
  We propose a new cooperative learning framework with pre-trained heterogeneous observation representations. We employ an encoder-decoder based graph attention to learn the intricate interactions and heterogeneous representations.
  arXiv  Detail & Related papers  (2020-12-24T04:52:29Z)
• Dif-MAML: Decentralized Multi-Agent Meta-Learning [54.39661018886268]
  We propose a cooperative multi-agent meta-learning algorithm, referred to as MAML or Dif-MAML. We show that the proposed strategy allows a collection of agents to attain agreement at a linear rate and to converge to a stationary point of the aggregate MAML. Simulation results illustrate the theoretical findings and the superior performance relative to the traditional non-cooperative setting.
  arXiv  Detail & Related papers  (2020-10-06T16:51:09Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.