mlsanityData to 7 Oct 2026

Which Agent Causes Task Failures and When? On Automated Failure Attribution of LLM Multi-Agent Systems

arXiv 2505.00212 ↗First posted 30 Apr 2025cs.MA cs.CLcited by 170

On the map: LLM agents and tool use → Software engineering coding agents

Abstract (arXiv)

Failure attribution in LLM multi-agent systems-identifying the agent and step responsible for task failures-provides crucial clues for systems debugging but remains underexplored and labor-intensive. In this paper, we propose and formulate a new research area: automated failure attribution for LLM multi-agent systems. To support this initiative, we introduce the Who&When dataset, comprising extensive failure logs from 127 LLM multi-agent systems with fine-grained annotations linking failures to specific agents and decisive error steps. Using the Who&When, 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, with some methods performing below random. Even SOTA reasoning models, such as OpenAI o1 and DeepSeek R1, fail to achieve practical usability. These results highlight the task's complexity and the need for further research in this area. Code and dataset are available at https://github.com/mingyin1/Agents_Failure_Attribution

Key facts. Cited by 170 papers in the map (data through 7 Oct 2026). Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use. Builds on 42 of its 51 references that are in the map. First posted on arXiv 30 Apr 2025.

Builds on

42 of 51 references.

References · 42

  1. 2023GPT-4 Technical Reportcited by 18,132
  2. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  3. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,393
  4. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,355
  5. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,146
  6. 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,064
  7. 2023Let's Verify Step by Stepcited by 3,598
  8. 2023SWE-bench: Can Language Models Resolve Real-World GitHub Issues?cited by 3,193
  9. 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Modelscited by 3,155
  10. 2023G-Eval: NLG Evaluation using GPT-4 with Better Human Alignmentcited by 2,205
  11. 2023MetaGPT: Meta Programming for A Multi-Agent Collaborative Frameworkcited by 1,761
  12. 2024A Survey on LLM-as-a-Judgecited by 1,384
30 more
  1. 2023CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Societycited by 1,348
  2. 2023Mind2Web: Towards a Generalist Agent for the Webcited by 1,210
  3. 2024OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environmentscited by 1,134
  4. 2023GAIA: a benchmark for General AI Assistantscited by 1,084
  5. 2023Large Language Models Cannot Self-Correct Reasoning Yetcited by 965
  6. 2023Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human Annotationscited by 895
  7. 2023ChatEval: Towards Better LLM-based Evaluators through Multi-Agent Debatecited by 724
  8. 2022Towards Reasoning in Large Language Models: A Surveycited by 593
  9. 2024From Generation to Judgment: Opportunities and Challenges of LLM-as-a-judgecited by 480
  10. 2025Process Reinforcement through Implicit Rewardscited by 478
  11. 2023GPTScore: Evaluate as You Desirecited by 357
  12. 2024JudgeBench: A Benchmark for Evaluating LLM-based Judgescited by 294
  13. 2024ProcessBench: Identifying Process Errors in Mathematical Reasoningcited by 277
  14. 2024Aguvis: Unified Pure Vision Agents for Autonomous GUI Interactioncited by 271
  15. 2024Agent-as-a-Judge: Evaluate Agents with Agentscited by 223
  16. 2024Magentic-One: A Generalist Multi-Agent System for Solving Complex Taskscited by 221
  17. 2024Judging the Judges: Evaluating Alignment and Vulnerabilities in LLMs-as-Judgescited by 197
  18. 2023SelfCheck: Using LLMs to Zero-Shot Check Their Own Step-by-Step Reasoningcited by 157
  19. 2024AssistantBench: Can Web Agents Solve Realistic and Time-Consuming Tasks?cited by 155
  20. 2024A Comprehensive Survey of Small Language Models in the Era of Large Language Models: Techniques, Enhancements, Applications, Collaboration with LLMs, and Trustworthinesscited by 125
  21. 2020Hyper-Parameter Optimization: A Review of Algorithms and Applicationscited by 114
  22. 2023Evil Geniuses: Delving into the Safety of LLM-based Agentscited by 103
  23. 2024SciAgents: Automating scientific discovery through multi-agent intelligent graph reasoningcited by 90
  24. 2025A Comprehensive Survey of Reward Models: Taxonomy, Applications, Challenges, and Futurecited by 53
  25. 2023EcoAssistant: Using LLM Assistant More Affordably and Accuratelycited by 47
  26. 2024Offline Training of Language Model Agents with Functions as Learnable Weightscited by 47
  27. 2024Adaptive In-conversation Team Building for Language Model Agentscited by 38
  28. 2024Needle in the Haystack for Memory Based Large Language Modelscited by 32
  29. 2024Towards Acyclic Preference Evaluation of Language Models via Multiple Evaluatorscited by 13
  30. 2024EcoAct: Economic Agent Determines When to Register What Actioncited by 11

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 29; 2026: 141.

170 papers.

Sort:

By area:

  1. 2025Where LLM Agents Fail and How They can Learn From Failurescited by 118
  2. 2025AgenTracer: Who Is Inducing Failure in the LLM Agentic Systems?cited by 105
  3. 2025DeepScientist: Advancing Frontier-Pushing Scientific Findings Progressivelycited by 56
  4. 2026AgentRx: Diagnosing AI Agent Failures from Execution Trajectoriescited by 53
  5. 2026Code as Agent Harnesscited by 40
  6. 2025CoAct-1: Computer-using Multi-Agent System with Coding Actionscited by 33
  7. 2025Who is Introducing the Failure? Automatically Attributing Failures of Multi-Agent Systems via Spectrum Analysiscited by 33
  8. 2025Single-agent or Multi-agent Systems? Why Not Both?cited by 28
  9. 2025Where Did It All Go Wrong? A Hierarchical Look into Multi-Agent Error Attributioncited by 25
  10. 2025DoVer: Intervention-Driven Auto Debugging for LLM Multi-Agent Systemscited by 23
  11. 2025SABER: Small Actions, Big Errors -- Safeguarding Mutating Steps in LLM Agentscited by 21
  12. 2026From Failed Trajectories to Reliable LLM Agents: Diagnosing and Repairing Harness Flawscited by 21
38 more
  1. 2025Beyond the Final Answer: Evaluating the Reasoning Trajectories of Tool-Augmented Agentscited by 20
  2. 2025RAFFLES: Reasoning-based Attribution of Faults for LLM Systemscited by 19
  3. 2025Aegis: Automated Error Generation and Attribution for Multi-Agent Systemscited by 19
  4. 2025Failure-Driven Workflow Refinementcited by 19
  5. 2026Flow caching for autoregressive video generationcited by 19
  6. 2026Causal Agent Replay: Counterfactual Attribution for LLM-Agent Failurescited by 18
  7. 2026CausalFlow: Causal Attribution and Counterfactual Repair for LLM Agent Failurescited by 17
  8. 2025LiveMCP-101: Stress Testing and Diagnosing MCP-enabled Agents on Challenging Queriescited by 15
  9. 2026AgentHallu: Benchmarking Automated Hallucination Attribution of LLM-based Agentscited by 15
  10. 2025MermaidFlow: Redefining Agentic Workflow Generation via Safety-Constrained Evolutionary Programmingcited by 14
  11. 2025Abduct, Act, Predict: Scaffolding Causal Inference for Automated Failure Attribution in Multi-Agent Systemscited by 14
  12. 2025CORRECT: COndensed eRror RECognition via knowledge Transfer in multi-agent systemscited by 14
  13. 2026From Spark to Fire: Modeling and Mitigating Error Cascades in LLM-Based Multi-Agent Collaborationcited by 13
  14. 2025SHIELDA: Structured Handling of Exceptions in LLM-Driven Agentic Workflowscited by 12
  15. 2026Agentic Confidence Calibrationcited by 12
  16. 2025Stop Wasting Your Tokens: Towards Efficient Runtime Multi-Agent Systemscited by 11
  17. 2026TrajAudit: Automated Failure Diagnosis for Agentic Coding Systemscited by 11
  18. 2026SkillAxe: Sharpening LLM-Authored Agent Skills Through Evaluation-Guided Self-Refinementcited by 11
  19. 2026Failure as a Process: An Anatomy of CLI Coding Agent Trajectoriescited by 11
  20. 2026Who&When Pro: Can LLMs Really Attribute Failures in AI Agents?cited by 11
  21. 2025Traceability and Accountability in Role-Specialized Multi-Agent LLM Pipelinescited by 10
  22. 2026From Confident Closing to Silent Failure: Characterizing False Success in LLM Agentscited by 10
  23. 2025Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agentscited by 9
  24. 2026FALAT: Tracing Failures in LLM Agent Trajectories via Dependency-Guided Searchcited by 9
  25. 2026AutoSaddler: Automatic Harness Optimization with Durable Updates from Agent Execution Tracescited by 9
  26. 2026The Why Behind the Action: Unveiling Internal Drivers via Agentic Attributioncited by 8
  27. 2026AgentForesight: Online Auditing for Early Failure Prediction in Multi-Agent Systemscited by 8
  28. 2025Beyond Task Completion: An Assessment Framework for Evaluating Agentic AI Systemscited by 7
  29. 2026Learning Decentralized LLM Collaboration with Multi-Agent Actor Criticcited by 7
  30. 2026Workspace Optimization: How to Train Your Agentcited by 7
  31. 2026Your Agents Are Aging Too: Agent Lifespan Engineering for Deployed Systemscited by 7
  32. 2025Agentic Services Computingcited by 6
  33. 2025Metacognitive Self-Correction for Multi-Agent System via Prototype-Guided Next-Execution Reconstructioncited by 6
  34. 2026Where Do Deep-Research Agents Go Wrong? Span-Level Error Localization in Agent Trajectoriescited by 6
  35. 2026Autonomous Research Agents: A Survey of AI Scientists and the Verification Gapcited by 6
  36. 2026Why Your Deep Research Agent Fails? On Hallucination Evaluation in Full Research Trajectorycited by 5
  37. 2026VerifyMAS: Hypothesis Verification for Failure Attribution in LLM Multi-Agent Systemscited by 5
  38. 2026MemTrace: Tracing and Attributing Errors in Large Language Model Memory Systemscited by 5

50 most cited of 170 All 170 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026Rethinking Failure Attribution in Multi-Agent Systems: A Multi-Perspective Benchmark and Evaluation
  2. 2026Seeing the Whole Elephant: A Benchmark for Failure Attribution in LLM-based Multi-Agent Systems
  3. 2026Adaptive Influence Graphs for Failure Attribution in Multi-Agent Systems
  4. 2026StepFinder: A Temporal Semantic Framework for Failure Attribution in Multi-Agent Systems
  5. 2026MASPrism: Lightweight Failure Attribution for Multi-Agent Systems Using Prefill-Stage Signals
  6. 2026AgentFixer: From Failure Detection to Fix Recommendations in LLM Agentic Systems
  7. 2026From Flat Logs to Causal Graphs: Hierarchical Failure Attribution for LLM-based Multi-Agent Systems
  8. 2026DCFA: Dual-view Causal-inspired Attribution for Failure Reasoning in LLM-based Multi-agent Systems