A survey that unifies prior work on multi-agent LLM systems via the LIFE framework, mapping dependencies across collaboration, failure attribution, and autonomous self-evolution while identifying cross-stage challenges.
hub Canonical reference
Beyond Self-Talk: A Communication-Centric Survey of LLM-Based Multi-Agent S ystems
Canonical reference. 100% of citing Pith papers cite this work as background.
22
Pith papers citing it
9
external citations · Pith
Background
100% of classified citations
abstract
Large language model-based multi-agent systems have recently gained significant attention due to their potential for complex, collaborative, and intelligent problem-solving capabilities. Existing surveys typically categorize LLM-based multi-agent systems (LLM-MAS) according to their application domains or architectures, overlooking the central role of communication in coordinating agent behaviors and interactions. To address this gap, this paper presents a comprehensive survey of LLM-MAS from a communication-centric perspective. Specifically, we propose a structured framework that integrates system-level communication (architecture, goals, and protocols) with system internal communication (strategies, paradigms, objects, and content), enabling a detailed exploration of how agents interact, negotiate, and achieve collective intelligence. Through an extensive analysis of recent literature, we identify key components in multiple dimensions and summarize their strengths and limitations. In addition, we highlight current challenges, including communication efficiency, security vulnerabilities, inadequate benchmarking, and scalability issues, and outline promising future research directions. This review aims to help researchers and practitioners gain a clear understanding of the communication mechanisms in LLM-MAS, thereby facilitating the design and deployment of robust, scalable, and secure multi-agent systems.
hub tools
citation-role summary
background 7
citation-polarity summary
roles
background 6polarities
background 6representative citing papers
PropGuard is a propagation-aware framework for LLM-MAS that constructs dual-view spatio-temporal graphs, employs a GE-GRPO inspector to recover suspicious subgraphs, and applies source-guided remediation to lower attack success while preserving task performance.
Anonymization in multi-agent debate reduces identity bias by equalizing self and peer weights in a Bayesian update model, quantified by the Identity Bias Coefficient.
A survey that defines Compound AI Systems, proposes a multi-dimensional taxonomy based on component roles and orchestration strategies, reviews four foundational paradigms, and identifies key challenges for future research.
A benchmark for LLM agents in partially observable joint decision-making reveals that deliberation challenges current models but can enable reflection and error correction.
PEAR is a permutation-equivariant adaptive routing protocol for multi-agent LLM debate that reconfigures sparse topologies each round to improve accuracy over fixed debate baselines.
Introduces EPC-AW to mitigate epistemic miscalibration in LLM multi-agent planning via consistency-based selection and refinement, reporting 9.75% average success improvement.
MASPO jointly optimizes prompts in multi-agent LLM systems via downstream-success evaluation and evolutionary beam search, delivering 2.9 average accuracy gains over prior methods across six tasks.
NOMAD decomposes UML class diagram creation into a multi-agent LLM workflow that outperforms baselines on a Northwind case study and human exercises while introducing a taxonomy of structural, relationship, and semantic errors.
A hierarchical multi-agent system with File-as-Bus durable state beats matched baselines on PaperBench and MLE-Bench Lite, and ablations show project-state continuity drives later-round gains.
Creates a five-dimension taxonomy (counterparty, payload, interaction state, discovery mechanism, schema flexibility) from nine protocols and identifies architectural patterns plus convergence trends.
A joint media selection and resource allocation algorithm (JMSRA) adaptively chooses token or KV-cache transmission and bandwidth allocation to reduce E2E latency compared to fixed baselines in wireless multi-agent systems.
Eywa enables language-based agentic AI systems to collaborate with specialized scientific foundation models for improved performance on structured data tasks.
WebMAC uses three specialized multi-agent modules to clarify test scenarios, partition them for adequacy, and generate executable scripts, yielding 30-60% higher success rates and 29% better efficiency than SOTA on four web systems.
The paper identifies twelve protocol-level security risks across MCP, A2A, Agora, and ANP and quantifies wrong-provider tool execution risk in MCP via a measurement-driven case study on multi-server composition.
Proposes a cross-layer intellicise network architecture grounded in multiple theories to support intelligent complex systems, with reviews of enabling technologies and a case study.
A Layer Execution Graph multi-agent system for hydrodynamics achieves 93.6% factual precision and 100% pass rate on 37 queries while degrading gracefully under data loss.
A survey comparing classical multi-agent systems with large foundation model-enabled multi-agent systems, showing how the latter enables semantic-level collaboration and greater adaptability.
The authors introduce agentic microphysics and generative safety to link local agent interactions to population-level risks in agentic AI through a causally explicit framework.
A survey consolidating benchmarks, agent frameworks, real-world applications, and protocols for LLM-based autonomous agents into a proposed taxonomy with recommendations for future research.
A survey of classical software design patterns applied to communication challenges in LLM agent systems using the Model Context Protocol.
citing papers explorer
-
Beyond Individual Intelligence: Surveying Collaboration, Failure Attribution, and Self-Evolution in LLM-based Multi-Agent Systems
A survey that unifies prior work on multi-agent LLM systems via the LIFE framework, mapping dependencies across collaboration, failure attribution, and autonomous self-evolution while identifying cross-stage challenges.
-
PropGuard: Safeguarding LLM-MAS via Propagation-Aware Exploration and Remediation
PropGuard is a propagation-aware framework for LLM-MAS that constructs dual-view spatio-temporal graphs, employs a GE-GRPO inspector to recover suspicious subgraphs, and applies source-guided remediation to lower attack success while preserving task performance.
-
When Identity Skews Debate: Anonymization for Bias-Reduced Multi-Agent Reasoning
Anonymization in multi-agent debate reduces identity bias by equalizing self and peer weights in a Bayesian update model, quantified by the Identity Bias Coefficient.
-
From Standalone LLMs to Integrated Intelligence: A Survey of Compound Al Systems
A survey that defines Compound AI Systems, proposes a multi-dimensional taxonomy based on component roles and orchestration strategies, reviews four foundational paradigms, and identifies key challenges for future research.
-
LLM Agents for Deliberative Collaboration: A Study on Joint Decision Making Under Partial Observability
A benchmark for LLM agents in partially observable joint decision-making reveals that deliberation challenges current models but can enable reflection and error correction.
-
PEAR: Permutation-Equivariant Adaptive Routing Multi-Agent Debate
PEAR is a permutation-equivariant adaptive routing protocol for multi-agent LLM debate that reconfigures sparse topologies each round to improve accuracy over fixed debate baselines.
-
When Planning Fails Despite Correct Execution: On Epistemic Calibration for LLM-Based Multi-Agent Systems
Introduces EPC-AW to mitigate epistemic miscalibration in LLM multi-agent planning via consistency-based selection and refinement, reporting 9.75% average success improvement.
-
MASPO: Joint Prompt Optimization for LLM-based Multi-Agent Systems
MASPO jointly optimizes prompts in multi-agent LLM systems via downstream-success evaluation and evolutionary beam search, delivering 2.9 average accuracy gains over prior methods across six tasks.
-
NOMAD: A Multi-Agent LLM System for UML Class Diagram Generation from Natural Language Requirements
NOMAD decomposes UML class diagram creation into a multi-agent LLM workflow that outperforms baselines on a Northwind case study and human exercises while introducing a taxonomy of structural, relationship, and semantic errors.
-
Toward Autonomous Long-Horizon Engineering for ML Research
A hierarchical multi-agent system with File-as-Bus durable state beats matched baselines on PaperBench and MLE-Bench Lite, and ablations show project-state continuity drives later-round gains.
-
A Technical Taxonomy of LLM Agent Communication Protocols
Creates a five-dimension taxonomy (counterparty, payload, interaction state, discovery mechanism, schema flexibility) from nine protocols and identifies architectural patterns plus convergence trends.
-
A Token/KV-Cache Communication Media Selection and Resource Allocation Strategy for Multi-Agent Collaboration
A joint media selection and resource allocation algorithm (JMSRA) adaptively chooses token or KV-cache transmission and bandwidth allocation to reduce E2E latency compared to fixed baselines in wireless multi-agent systems.
-
Heterogeneous Scientific Foundation Model Collaboration
Eywa enables language-based agentic AI systems to collaborate with specialized scientific foundation models for improved performance on structured data tasks.
-
WebMAC: A Multi-Agent Collaborative Framework for Scenario Testing of Web Systems
WebMAC uses three specialized multi-agent modules to clarify test scenarios, partition them for adequacy, and generate executable scripts, yielding 30-60% higher success rates and 29% better efficiency than SOTA on four web systems.
-
Security Threat Modeling for Emerging AI-Agent Protocols: A Comparative Analysis of MCP, A2A, Agora, and ANP
The paper identifies twelve protocol-level security risks across MCP, A2A, Agora, and ANP and quantifies wrong-provider tool execution risk in MCP via a measurement-driven case study on multi-server composition.
-
Evolving Intelligent Complex Systems via Intellicise Networks: Architecture, Technologies, and Pathways
Proposes a cross-layer intellicise network architecture grounded in multiple theories to support intelligent complex systems, with reviews of enabling technologies and a case study.
-
Towards Multi-Agent Autonomous Reasoning in Hydrodynamics
A Layer Execution Graph multi-agent system for hydrodynamics achieves 93.6% factual precision and 100% pass rate on 37 queries while degrading gracefully under data loss.
-
Multi-Agent Systems: From Classical Paradigms to Large Foundation Model-Enabled Futures
A survey comparing classical multi-agent systems with large foundation model-enabled multi-agent systems, showing how the latter enables semantic-level collaboration and greater adaptability.
-
Agentic Microphysics: A Manifesto for Generative AI Safety
The authors introduce agentic microphysics and generative safety to link local agent interactions to population-level risks in agentic AI through a causally explicit framework.
-
From LLM Reasoning to Autonomous AI Agents: A Comprehensive Review
A survey consolidating benchmarks, agent frameworks, real-world applications, and protocols for LLM-based autonomous agents into a proposed taxonomy with recommendations for future research.
-
Survey of LLM Agent Communication with MCP: A Software Design Pattern Centric Review
A survey of classical software design patterns applied to communication challenges in LLM agent systems using the Model Context Protocol.
- Learning to Interrupt in Language-based Multi-agent Communication