Why Do Multi-Agent LLM Systems Fail?

• URL: http://arxiv.org/abs/2503.13657v3
• Date: Sun, 26 Oct 2025 23:25:18 GMT
• Title: Why Do Multi-Agent LLM Systems Fail?
• Authors: Mert Cemri, Melissa Z. Pan, Shuyi Yang, Lakshya A. Agrawal, Bhavya Chopra, Rishabh Tiwari, Kurt Keutzer, Aditya Parameswaran, Dan Klein, Kannan Ramchandran, Matei Zaharia, Joseph E. Gonzalez, Ion Stoica,
• Abstract summary: We introduce MAST-Data, a comprehensive dataset of 1600+ annotated traces collected across 7 popular MAS frameworks.<n>We build the first Multi-Agent System Failure taxonomy (MAST)<n>We leverage MAST and MAST-Data to analyze failure patterns across models (GPT4, Claude 3, Qwen2.5, CodeLlama) and tasks (coding, math, general agent)
• Score: 87.90075668488434
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Despite enthusiasm for Multi-Agent LLM Systems (MAS), their performance gains on popular benchmarks are often minimal. This gap highlights a critical need for a principled understanding of why MAS fail. Addressing this question requires systematic identification and analysis of failure patterns. We introduce MAST-Data, a comprehensive dataset of 1600+ annotated traces collected across 7 popular MAS frameworks. MAST-Data is the first multi-agent system dataset to outline the failure dynamics in MAS for guiding the development of better future systems. To enable systematic classification of failures for MAST-Data, we build the first Multi-Agent System Failure Taxonomy (MAST). We develop MAST through rigorous analysis of 150 traces, guided closely by expert human annotators and validated by high inter-annotator agreement (kappa = 0.88). This process identifies 14 unique modes, clustered into 3 categories: (i) system design issues, (ii) inter-agent misalignment, and (iii) task verification. To enable scalable annotation, we develop an LLM-as-a-Judge pipeline with high agreement with human annotations. We leverage MAST and MAST-Data to analyze failure patterns across models (GPT4, Claude 3, Qwen2.5, CodeLlama) and tasks (coding, math, general agent), demonstrating improvement headrooms from better MAS design. Our analysis provides insights revealing that identified failures require more sophisticated solutions, highlighting a clear roadmap for future research. We publicly release our comprehensive dataset (MAST-Data), the MAST, and our LLM annotator to facilitate widespread research and development in MAS.

Related papers

• Process-Centric Analysis of Agentic Software Systems [10.976178600911263]
  We introduce Graphectory to encode the temporal and semantic relations in software systems.<n>We analyze 4000 trajectories of two dominant agentic programming models.<n>Our fully automated analyses reveal that agents using richer prompts exhibit more complex Graphectory.
  arXiv  Detail & Related papers  (2025-12-02T04:12:29Z)
• Metacognitive Self-Correction for Multi-Agent System via Prototype-Guided Next-Execution Reconstruction [58.51530390018909]
  Large Language Model based multi-agent systems excel at collaborative problem solving but remain brittle to cascading errors.<n>We present MASC, a metacognitive framework that endows MAS with real-time, unsupervised, step-level error detection and self-correction.
  arXiv  Detail & Related papers  (2025-10-16T05:35:37Z)
• Testing and Enhancing Multi-Agent Systems for Robust Code Generation [21.38351747327572]
  Multi-agent systems (MASs) have emerged as a promising paradigm for automated code generation.<n>Despite their prosperous development and adoption, their robustness remains pressingly under-explored.<n>This paper presents the first comprehensive study examining the robustness of MASs for code generation through a fuzzing-based testing approach.
  arXiv  Detail & Related papers  (2025-10-12T05:45:04Z)
• CoDA: Agentic Systems for Collaborative Data Visualization [57.270599188947294]
  Deep research has revolutionized data analysis, yet data scientists still devote substantial time to manually crafting visualizations.<n>Existing approaches, including simple single- or multi-agent systems, often oversimplify the task.<n>We introduce CoDA, a multi-agent system that employs specialized LLM agents for metadata analysis, task planning, code generation, and self-reflection.
  arXiv  Detail & Related papers  (2025-10-03T17:30:16Z)
• Where LLM Agents Fail and How They can Learn From Failures [62.196870049524364]
  Large Language Model (LLM) agents have shown promise in solving complex, multi-step tasks.<n>They amplify vulnerability to cascading failures, where a single root-cause error propagates through subsequent decisions.<n>Current systems lack a framework that can comprehensively understand agent error in a modular and systemic way.<n>We introduce the AgentErrorTaxonomy, a modular classification of failure modes spanning memory, reflection, planning, action, and system-level operations.
  arXiv  Detail & Related papers  (2025-09-29T18:20:27Z)
• LumiMAS: A Comprehensive Framework for Real-Time Monitoring and Enhanced Observability in Multi-Agent Systems [20.35152929165441]
  The proposed framework consists of three key components: a monitoring and logging layer, anomaly detection layer, and anomaly explanation layer.<n>LumiMAS was evaluated on seven different MAS applications, implemented using two popular MAS platforms, and a diverse set of possible failures.
  arXiv  Detail & Related papers  (2025-08-17T15:55:02Z)
• MCP-Orchestrated Multi-Agent System for Automated Disinformation Detection [84.75972919995398]
  This paper presents a multi-agent system that uses relation extraction to detect disinformation in news articles.<n>The proposed Agentic AI system combines four agents: (i) a machine learning agent (logistic regression), (ii) a Wikipedia knowledge check agent, and (iv) a web-scraped data analyzer.<n>Results demonstrate that the multi-agent ensemble achieves 95.3% accuracy with an F1 score of 0.964, significantly outperforming individual agents and traditional approaches.
  arXiv  Detail & Related papers  (2025-08-13T19:14:48Z)
• Which Agent Causes Task Failures and When? On Automated Failure Attribution of LLM Multi-Agent Systems [50.29939179830491]
  Failure attribution in LLM multi-agent systems remains underexplored and labor-intensive. We develop and evaluate three automated failure attribution methods, summarizing their corresponding pros and cons. The best method achieves 53.5% accuracy in identifying failure-responsible agents but only 14.2% in pinpointing failure steps.
  arXiv  Detail & Related papers  (2025-04-30T23:09:44Z)
• Grounded Chain-of-Thought for Multimodal Large Language Models [66.04061083611863]
  We propose a new learning task for multimodal large language models (MLLMs) called Grounded Chain-of-Thought (GCoT) GCoT is keen to helping MLLMs to recognize and ground the relevant visual cues step by step, thereby predicting the correct answer with grounding coordinates as the intuitive basis. To facilitate this task, we also carefully design and construct a dataset called multimodal grounded chain-of-thought (MM-GCoT) consisting of 24,022 GCoT examples for 5,033 images.
  arXiv  Detail & Related papers  (2025-03-17T04:07:47Z)
• OR-LLM-Agent: Automating Modeling and Solving of Operations Research Optimization Problems with Reasoning LLM [15.260794368585692]
  We propose OR-LLM-Agent, an AI agent framework built on reasoning LLMs for automated Operations Research problem solving.<n>We show that OR-LLM-Agent utilizing DeepSeek-R1 in its framework outperforms advanced methods, including GPT-o3, Gemini 2.5 Pro, DeepSeek-R1, and ORLM, by at least 7% in accuracy.
  arXiv  Detail & Related papers  (2025-03-13T03:40:50Z)
• ReMA: Learning to Meta-think for LLMs with Multi-Agent Reinforcement Learning [53.817538122688944]
  We introduce Reinforced Meta-thinking Agents (ReMA) to elicit meta-thinking behaviors from Reasoning of Large Language Models (LLMs)<n>ReMA decouples the reasoning process into two hierarchical agents: a high-level meta-thinking agent responsible for generating strategic oversight and plans, and a low-level reasoning agent for detailed executions.<n> Empirical results from single-turn experiments demonstrate that ReMA outperforms single-agent RL baselines on complex reasoning tasks.
  arXiv  Detail & Related papers  (2025-03-12T16:05:31Z)
• VOILA: Evaluation of MLLMs For Perceptual Understanding and Analogical Reasoning [63.0285363282581]
  Multimodal Large Language Models (MLLMs) have become a powerful tool for integrating visual and textual information. We introduce VOILA, a benchmark designed to evaluate MLLMs' perceptual understanding and abstract relational reasoning. We reveal that current MLLMs struggle to comprehend inter-image relationships and exhibit limited capabilities in high-level relational reasoning.
  arXiv  Detail & Related papers  (2025-02-25T23:36:19Z)
• Investigating the Adaptive Robustness with Knowledge Conflicts in LLM-based Multi-Agent Systems [39.390472904456836]
  We design four comprehensive metrics to investigate robustness of multi-agent systems (MASs) We first analyze mild knowledge conflicts introduced by heterogeneous agents and find that they do not harm system robustness but instead improve collaborative decision-making. Finally, we conduct ablation studies on the knowledge conflict number, agent number, and interaction rounds, finding that the self-repairing capability of MASs has intrinsic limits.
  arXiv  Detail & Related papers  (2025-02-21T02:24:43Z)
• D-CIPHER: Dynamic Collaborative Intelligent Agents with Planning and Heterogeneous Execution for Enhanced Reasoning in Offensive Security [22.86304661035188]
  Large Language Models (LLMs) have been used in cybersecurity in many ways.<n>Capture the Flag (CTF) challenges serve as benchmarks for assessing the automated task-planning abilities of LLM agents.<n>We introduce the D-CIPHER multi-agent LLM framework for collaborative CTF challenge solving.
  arXiv  Detail & Related papers  (2025-02-15T23:43:18Z)
• MASTER: A Multi-Agent System with LLM Specialized MCTS [11.780059513577848]
  Large Language Models (LLM) are increasingly being explored for problem-solving tasks. MCTS relies on extensive sampling simulations to approximate the true reward distribution. We present a novel framework that coordinates agent recruitment and communication through LLM specialized MCTS.
  arXiv  Detail & Related papers  (2025-01-24T08:01:11Z)
• MALT: Improving Reasoning with Multi-Agent LLM Training [66.9481561915524]
  MALT (Multi-Agent LLM Training) is a novel post-training strategy that divides the reasoning process into generation, verification, and refinement steps.<n>On MATH, GSM8K, and CSQA, MALT surpasses the same baseline LLM with a relative improvement of 15.66%, 7.42%, and 9.40% respectively.
  arXiv  Detail & Related papers  (2024-12-02T19:30:36Z)
• Enhancing Multi-Agent Consensus through Third-Party LLM Integration: Analyzing Uncertainty and Mitigating Hallucinations in Large Language Models [1.4582633500696451]
  Large Language Models (LLMs) still face challenges when dealing with complex reasoning tasks. This paper proposes a new method that integrates different LLMs to expand the knowledge boundary. Experiments on arithmetic datasets have validated the effectiveness of the method.
  arXiv  Detail & Related papers  (2024-11-25T08:42:33Z)
• ErrorRadar: Benchmarking Complex Mathematical Reasoning of Multimodal Large Language Models Via Error Detection [60.297079601066784]
  We introduce ErrorRadar, the first benchmark designed to assess MLLMs' capabilities in error detection. ErrorRadar evaluates two sub-tasks: error step identification and error categorization. It consists of 2,500 high-quality multimodal K-12 mathematical problems, collected from real-world student interactions. Results indicate significant challenges still remain, as GPT-4o with best performance is still around 10% behind human evaluation.
  arXiv  Detail & Related papers  (2024-10-06T14:59:09Z)
• Textualized Agent-Style Reasoning for Complex Tasks by Multiple Round LLM Generation [49.27250832754313]
  We present AgentCOT, a llm-based autonomous agent framework. At each step, AgentCOT selects an action and executes it to yield an intermediate result with supporting evidence. We introduce two new strategies to enhance the performance of AgentCOT.
  arXiv  Detail & Related papers  (2024-09-19T02:20:06Z)
• SUPER: Evaluating Agents on Setting Up and Executing Tasks from Research Repositories [55.161075901665946]
  Super aims to capture the realistic challenges faced by researchers working with Machine Learning (ML) and Natural Language Processing (NLP) research repositories. Our benchmark comprises three distinct problem sets: 45 end-to-end problems with annotated expert solutions, 152 sub problems derived from the expert set that focus on specific challenges, and 602 automatically generated problems for larger-scale development. We show that state-of-the-art approaches struggle to solve these problems with the best model (GPT-4o) solving only 16.3% of the end-to-end set, and 46.1% of the scenarios.
  arXiv  Detail & Related papers  (2024-09-11T17:37:48Z)
• Visual Reasoning and Multi-Agent Approach in Multimodal Large Language Models (MLLMs): Solving TSP and mTSP Combinatorial Challenges [5.934258790280767]
  Multimodal Large Language Models (MLLMs) harness comprehensive knowledge spanning text, images, and audio to adeptly tackle complex problems. This study explores the ability of MLLMs in visually solving the Traveling Salesman Problem (TSP) and Multiple Traveling Salesman Problem (mTSP) We introduce a novel approach employing multiple specialized agents within the MLLM framework, each dedicated to optimizing solutions for these challenges.
  arXiv  Detail & Related papers  (2024-06-26T07:12:06Z)
• AutoDetect: Towards a Unified Framework for Automated Weakness Detection in Large Language Models [95.09157454599605]
  Large Language Models (LLMs) are becoming increasingly powerful, but they still exhibit significant but subtle weaknesses.<n>Traditional benchmarking approaches cannot thoroughly pinpoint specific model deficiencies.<n>We introduce a unified framework, AutoDetect, to automatically expose weaknesses in LLMs across various tasks.
  arXiv  Detail & Related papers  (2024-06-24T15:16:45Z)
• Smurfs: Leveraging Multiple Proficiency Agents with Context-Efficiency for Tool Planning [14.635361844362794]
  Smurfs' is a cutting-edge multi-agent framework designed to revolutionize the application of large language models. Smurfs can enhance the model's ability to solve complex tasks at no additional cost.
  arXiv  Detail & Related papers  (2024-05-09T17:49:04Z)

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.