mlsanityData to 7 Oct 2026

Understanding the Information Propagation Effects of Communication Topologies in LLM-based Multi-Agent Systems

arXiv 2505.23352 ↗First posted 29 May 2025cs.MA cs.AIcited by 39

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

Abstract (arXiv)

The communication topology in large language model-based multi-agent systems fundamentally governs inter-agent collaboration patterns, critically shaping both the efficiency and effectiveness of collective decision-making. While recent studies for communication topology automated design tend to construct sparse structures for efficiency, they often overlook why and when sparse and dense topologies help or hinder collaboration. In this paper, we present a causal framework to analyze how agent outputs, whether correct or erroneous, propagate under topologies with varying sparsity. Our empirical studies reveal that moderately sparse topologies, which effectively suppress error propagation while preserving beneficial information diffusion, typically achieve optimal task performance. Guided by this insight, we propose a novel topology design approach, EIB-leanrner, that balances error suppression and beneficial information propagation by fusing connectivity patterns from both dense and sparse graphs. Extensive experiments show the superior effectiveness, communication cost, and robustness of EIB-leanrner.

Key facts. Cited by 39 papers in the map (data through 7 Oct 2026). Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use. Builds on 24 of its 38 references that are in the map. First posted on arXiv 29 May 2025.

Builds on

24 of 38 references.

Closest ancestors · 11

Same direction, newest first

  1. 2025Advances and Challenges in Foundation Agents: From Brain-Inspired Intelligence to Evolutionary, Collaborative, and Safe Systems31 Mar
  2. 2025Multi-Agent Design: Optimizing Agents with Better Prompts and Topologies4 Feb
  3. 2024AgentGen: Enhancing Planning Abilities for Large Language Model based Agent via Environment and Task Generation1 Aug
  4. 2024Improving Multi-Agent Debate with Sparse Communication Topology17 Jun
  5. 2024Chain of Agents: Large Language Models Collaborating on Long-Context Tasks4 Jun
  6. 2023Language Agent Tree Search Unifies Reasoning Acting and Planning in Language Models6 Oct
  7. 2023The Rise and Potential of Large Language Model Based Agents: A Survey14 Sep
  8. 2023A Survey on Large Language Model based Autonomous Agents22 Aug
  9. 2023Unleashing the Emergent Cognitive Synergy in Large Language Models: A Task-Solving Agent through Multi-Persona Self-Collaboration11 Jul
  10. 2023Multi-Agent Collaboration: Harnessing the Power of Intelligent LLM Agents5 Jun
  11. 2023CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Society31 Mar

Other references · 13

  1. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,355
  2. 2020Measuring Massive Multitask Language Understandingcited by 7,768
  3. 2023Improving Factuality and Reasoning in Language Models through Multiagent Debatecited by 1,739
  4. 2023Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debatecited by 995
  5. 2021Are NLP Models really able to Solve Simple Math Word Problems?cited by 861
  6. 2017Program Induction by Rationale Generation : Learning to Solve and Explain Algebraic Word Problemscited by 642
  7. 2016Solving General Arithmetic Word Problemscited by 353
  8. 2024CodeAgent: Enhancing Code Generation with Tool-Integrated Agent Systems for Real-World Repo-level Coding Challengescited by 278
  9. 2023Exploring Collaboration Mechanisms for LLM Agents: A Social Psychology Viewcited by 221
  10. 2024A Survey on Recent Advances in LLM-Based Multi-turn Dialogue Systemscited by 170
  11. 2024Self-Organized Agents: A LLM Multi-Agent Framework toward Ultra Large-Scale Code Generation and Optimizationcited by 84
  12. 2024Learning Multi-Agent Communication from Graph Modeling Perspectivecited by 47
1 more
  1. 2025GraphMaster: Automated Graph Synthesis via LLM Agents in Data-Limited Environmentscited by 18

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 9; 2026: 30.

39 papers.

Sort:

By area:

  1. 2025FaithCoT-Bench: Benchmarking Instance-Level Faithfulness of Chain-of-Thought Reasoningcited by 25
  2. 2025A Survey of Generalization of Graph Anomaly Detection: From Transfer Learning to Foundation Modelscited by 13
  3. 2025Stop Wasting Your Tokens: Towards Efficient Runtime Multi-Agent Systemscited by 11
  4. 2026MAS-FIRE: Fault Injection and Reliability Evaluation for LLM-Based Multi-Agent Systemscited by 11
  5. 2025Explainable and Fine-Grained Safeguarding of LLM Multi-Agent Systems via Bi-Level Graph Anomaly Detectioncited by 8
  6. 2026Thought Virus: Viral Misalignment via Subliminal Prompting in Multi-Agent Systemscited by 7
  7. 2025Metacognitive Self-Correction for Multi-Agent System via Prototype-Guided Next-Execution Reconstructioncited by 6
  8. 2026GoAgent: Group-of-Agents Communication Topology Generation for LLM-based Multi-Agent Systemscited by 5
  9. 2026Graph-GRPO: Stabilizing Multi-Agent Topology Learning via Group Relative Policy Optimizationcited by 4
  10. 2026MAS-PromptBench: When Does Prompt Optimization Improve Multi-Agent LLM Systems?cited by 4
  11. 2025Don't Trust Your Upstream: Exploiting LLM Multi-Agent System via Topology-Guided Adversarial Propagationcited by 3
  12. 2025AgentBalance: Backbone-then-Topology Design for Cost-Effective Multi-Agent Systems under Budget Constraintscited by 3
27 more
  1. 2026Beyond a Single Perspective: Text Anomaly Detection with Multi-View Language Representationscited by 3
  2. 2026AgentCollabBench: Diagnosing When Good Agents Make Bad Collaboratorscited by 3
  3. 2026CASPIAN: Online Detection and Attribution of Cascade Attacks in LLM Multi-Agent Systems via Cross-Channel Causal Monitoringcited by 3
  4. 2026Towards Adaptive, Scalable, and Robust Coordination of LLM Agents: A Dynamic Ad-Hoc Networking Perspectivecited by 2
  5. 2026Experience as a Compass: Multi-agent RAG with Evolving Orchestration and Agent Promptscited by 2
  6. 2026Learning to Communicate: Toward End-to-End Optimization of Multi-Agent Language Systemscited by 2
  7. 2026QUIVER: A Formal Framework for Quantifying Perturbation Propagation and Bifurcation in Compound AI Systemscited by 2
  8. 2025Test-time GNN Model Evaluation on Dynamic Graphscited by 1
  9. 2025Dual Mamba for Node-Specific Representation Learning: Tackling Over-Smoothing with Selective State Space Modelingcited by 1
  10. 2026What Do Agents Communicate? Characterizing Information Exchange in Multi-Agent Systemscited by 1
  11. 2026Large language models for spreading dynamics in complex systemscited by 0
  12. 2026Active Learning for Communication Structure Optimization in LLM-Based Multi-Agent Systemscited by 0
  13. 2026MasFACT: Continual Multi-Agent Topology Learning via Geometry-Aware Posterior Transfercited by 0
  14. 2026Enhancing Multi-Agent Communication through Attention Steering with Context Relevancecited by 0
  15. 2026When Do Multi-Agent Systems Help? An Information Bottleneck Perspectivecited by 0
  16. 2026Reliability-Contagion Feasibility in LLM Multi-Agent Networkscited by 0
  17. 2026Attacking and Defending Multi-Agent Collaborative Filtering Systems Through Connectivitycited by 0
  18. 2026Interaction Creates Dynamical AI Behavior Absent in Isolationcited by 0
  19. 2026ForestBench: A Unified Graph Framework for Evaluating Multi-Agent Collaborationcited by 0
  20. 2026Building Trustworthy Graph-Agentic RAG for Social Good: Architectures, Failure Propagation, and Assurance by Constructioncited by 0
  21. 2026Translating the Translator: Decomposing the Cost of English-Forced Inter-Agent Communicationcited by 0
  22. 2026Decomposition Buys Integrity, Not Yieldcited by 0
  23. 2026Proxifield: Decentralized Multi-Agent Communication through Semantic Proximitycited by 0
  24. 2026OpenMAS-GCom. A Diagnostic Benchmark for Graph-enhanced Multi-Agent Systemscited by 0
  25. 2026MAGIC: Mixed-Granularity Agent Graphs via Incremental Construction with Dense-Reward Reinforcement Learningcited by 0
  26. 2026Do We Need Complex Topology Control? Distinct-Peer Random Routing Improves Cost-Efficiency in Sparse Multi-Agent Debatecited by 0
  27. 2026Social Circuits behind Multi-agent Echo Chamberscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Discovering Efficient and Explainable Communication Topologies for LLM-based Multi-Agent Systems via Causal Inference
  2. 2025Rethinking Multi-Agent Intelligence Through the Lens of Small-World Networks
  3. 2025Topological Structure Learning Should Be A Research Priority for LLM-Based Multi-Agent Systems
  4. 2025Dynamic Generation of Multi-LLM Agents Communication Topologies with Graph Diffusion Models
  5. 2026Conformity Dynamics in LLM Multi-Agent Systems: The Roles of Topology and Self-Social Weighting
  6. 2024Cut the Crap: An Economical Communication Pipeline for LLM-based Multi-Agent Systems
  7. 2024G-Designer: Architecting Multi-agent Communication Topologies via Graph Neural Networks
  8. 2026Codebook Agent: Amortized Topology Design for LLM Multi-Agent Systems