mlsanityData to 7 Oct 2026

What Do Agents Communicate? Characterizing Information Exchange in Multi-Agent Systems

arXiv 2605.20548 ↗First posted 19 May 2026cs.MAcited by 1

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

Abstract (arXiv)

Large Language Models (LLMs) have enabled collaborative Multi-Agent (MA) systems, where interacting agents improve performance through diverse reasoning and iterative refinement. However, these systems remain vulnerable to error propagation, where early-stage information degrades downstream reasoning. To address this, we conduct a systematic analysis of inter-agent communication to identify which information drives MA performance. We find that the absence of reasoning and verification in inter-agent communication significantly degrades performance. Based on these insights, we propose Category-Aware Recovery Augmentation (technique), which enforces the presence of critical information during communication. recovers up to 86.2% of failed cases. Our results highlight the key role of information quality in effective MA collaboration. Our code is available at https://anonymous.4open.science/r/cara_mas

Key facts. Cited by 1 paper 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 51 of its 61 references that are in the map. First posted on arXiv 19 May 2026.

Builds on

51 of 61 references.

Closest ancestors · 12

Same direction, newest first

  1. 2025Stop Wasting Your Tokens: Towards Efficient Runtime Multi-Agent Systems30 Oct
  2. 2025Understanding the Information Propagation Effects of Communication Topologies in LLM-based Multi-Agent Systems29 May
  3. 2025Multi-Agent Collaboration via Evolving Orchestration26 May
  4. 2025Multi-Agent Design: Optimizing Agents with Better Prompts and Topologies4 Feb
  5. 2024Cut the Crap: An Economical Communication Pipeline for LLM-based Multi-Agent Systems3 Oct
  6. 2024MAgICoRe: Multi-Agent, Iterative, Coarse-to-Fine Refinement for Reasoning18 Sep
  7. 2024Improving Multi-Agent Debate with Sparse Communication Topology17 Jun
  8. 2024Scaling Large Language Model-based Multi-Agent Collaboration11 Jun
  9. 2023Multi-Agent Consensus Seeking via Large Language Models31 Oct
  10. 2023ChatEval: Towards Better LLM-based Evaluators through Multi-Agent Debate14 Aug
  11. 2023Unleashing the Emergent Cognitive Synergy in Large Language Models: A Task-Solving Agent through Multi-Persona Self-Collaboration11 Jul
  12. 2023CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Society31 Mar

Other references · 39

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  2. 2021Training Verifiers to Solve Math Word Problemscited by 9,743
  3. 2025Qwen3 Technical Reportcited by 8,195
  4. 2020Measuring Massive Multitask Language Understandingcited by 7,768
  5. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,601
  6. 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,064
  7. 2023Let's Verify Step by Stepcited by 3,598
  8. 2023Generative Agents: Interactive Simulacra of Human Behaviorcited by 3,500
  9. 2024LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Codecited by 1,906
  10. 2023Improving Factuality and Reasoning in Language Models through Multiagent Debatecited by 1,739
  11. 2017Understanding Black-box Predictions via Influence Functionscited by 1,714
  12. 2023Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debatecited by 995
27 more
  1. 2023Large Language Models Cannot Self-Correct Reasoning Yetcited by 965
  2. 2019SpanBERT: Improving Pre-training by Representing and Predicting Spanscited by 832
  3. 2021Did Aristotle Use a Laptop? A Question Answering Benchmark with Implicit Reasoning Strategiescited by 736
  4. 2024AFlow: Automating Agentic Workflow Generationcited by 395
  5. 2024CRUXEval: A Benchmark for Code Reasoning, Understanding and Executioncited by 294
  6. 2024Large Language Models for Data Annotation and Synthesis: A Surveycited by 201
  7. 2024Large Language Model-Based Agents for Software Engineering: A Surveycited by 173
  8. 2024LLM-Based Multi-Agent Systems for Software Engineering: Literature Review, Vision and the Road Aheadcited by 155
  9. 2025Where LLM Agents Fail and How They can Learn From Failurescited by 118
  10. 2023Should we be going MAD? A Look at Multi-Agent Debate Strategies for LLMscited by 103
  11. 2023Exchange-of-Thought: Enhancing Large Language Model Capabilities through Cross-Model Communicationcited by 78
  12. 2025Debate or Vote: Which Yields Better Decisions in Multi-Agent Large Language Models?cited by 74
  13. 2025Evaluating Step-by-step Reasoning Traces: A Surveycited by 62
  14. 2025Voting or Consensus? Decision-Making in Multi-Agent Debatecited by 55
  15. 2024AgileCoder: Dynamic Collaborative Agents for Software Development based on Agile Methodologycited by 45
  16. 2024Can LLMs Replace Manual Annotation of Software Engineering Artifacts?cited by 40
  17. 2024Knowledge Boundary of Large Language Models: A Surveycited by 36
  18. 2023Open, Closed, or Small Language Models for Text Classification?cited by 34
  19. 2025Process-Centric Analysis of Agentic Software Systemscited by 24
  20. 2025Just Put a Human in the Loop? Investigating LLM-Assisted Annotation for Subjective Taskscited by 21
  21. 2025Unified Software Engineering Agent as AI Software Engineercited by 17
  22. 2020Considering Likelihood in NLP Classification Explanations with Occlusion and Language Modelingcited by 11
  23. 2025Enhancing LLM Performance Through Debate: An Empirical Study on Multi-Agent Debate for Coding Taskscited by 11
  24. 2026Wink: Recovering from Misbehaviors in Coding Agentscited by 11
  25. 2025iMAD: Intelligent Multi-Agent Debate for Efficient and Accurate LLM Inferencecited by 10
  26. 2025Can Agents Fix Agent Issues?cited by 8
  27. 2025An LLM-as-Judge Metric for Bridging the Gap with Human Evaluation in SE Taskscited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026CollabFlow: Recursive Self-Improvement of Agent Collaborationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Beyond Final Accuracy: Auditing Communication in LLM Multi-Agent Systems
  2. 2025Communication and Verification in LLM Agents towards Collaboration under Information Asymmetry
  3. 2026What Should Agents Say? Action-state Communication for Efficient Multi-Agent Systems
  4. 2026The Five Ws of Multi-Agent Communication: Who Talks to Whom, When, What, and Why -- A Survey from MARL to Emergent Language and LLMs
  5. 2025Systematic Failures in Collective Reasoning under Distributed Information in Multi-Agent LLMs
  6. 2026When Does Multi-Agent Collaboration Help? An Entropy Perspective
  7. 2024COMMA: A Communicative Multimodal Multi-Agent Benchmark
  8. 2026More Capable, Less Cooperative? When LLMs Fail At Zero-Cost Collaboration