mlsanityData to 7 Oct 2026

SWE-Debate: Competitive Multi-Agent Debate for Software Issue Resolution

arXiv 2507.23348 ↗First posted 31 Jul 2025cs.SE cs.CL cs.LGcited by 52

On the map: LLM reasoning → Reasoning for agents, planning and code

also in 1 more
Abstract (arXiv)

Issue resolution has made remarkable progress thanks to the advanced reasoning capabilities of large language models (LLMs). Recently, agent-based frameworks such as SWE-agent have further advanced this progress by enabling autonomous, tool-using agents to tackle complex software engineering tasks. While existing agent-based issue resolution approaches are primarily based on agents' independent explorations, they often get stuck in local solutions and fail to identify issue patterns that span across different parts of the codebase. To address this limitation, we propose SWE-Debate, a competitive multi-agent debate framework that encourages diverse reasoning paths and achieves more consolidated issue localization. SWE-Debate first creates multiple fault propagation traces as localization proposals by traversing a code dependency graph. Then, it organizes a three-round debate among specialized agents, each embodying distinct reasoning perspectives along the fault propagation trace. This structured competition enables agents to collaboratively converge on a consolidated fix plan. Finally, this consolidated fix plan is integrated into an MCTS-based code modification agent for patch generation. Experiments on the SWE-bench benchmark show that SWE-Debate achieves new state-of-the-art results in open-source agent frameworks and outperforms baselines by a large margin.

Key facts. Cited by 52 papers in the map (data through 7 Oct 2026). Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning. Builds on 29 of its 88 references that are in the map. First posted on arXiv 31 Jul 2025.

Builds on

29 of 88 references.

Closest ancestors · 1

Same direction, newest first

  1. 2025CodeMonkeys: Scaling Test-Time Compute for Software Engineering24 Jan

Other references · 28

  1. 2023SWE-bench: Can Language Models Resolve Real-World GitHub Issues?cited by 3,193
  2. 2023AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversationcited by 1,906
  3. 2023Improving Factuality and Reasoning in Language Models through Multiagent Debatecited by 1,739
  4. 2024OpenHands: An Open Platform for AI Software Developers as Generalist Agentscited by 956
  5. 2024Training Software Engineering Agents and Verifiers with SWE-Gymcited by 300
  6. 2024AutoCodeRover: Autonomous Program Improvementcited by 272
  7. 2025SWE-smith: Scaling Data for Software Engineering Agentscited by 255
  8. 2024SWT-Bench: Testing and Validating Real-World Bug-Fixes with Code Agentscited by 129
  9. 2024CodeR: Issue Resolving with Multi-Agent and Task Graphscited by 89
  10. 2025LocAgent: Graph-Guided LLM Agents for Code Localizationcited by 82
  11. 2025SWE-bench Goes Live!cited by 74
  12. 2024MarsCode Agent: AI-native Automated Bug Fixingcited by 68
16 more
  1. 2023Large Language Models for Test-Free Fault Localizationcited by 67
  2. 2023RCAgent: Cloud Root Cause Analysis by Autonomous Agents with Tool-Augmented Large Language Modelscited by 62
  3. 2024Alibaba LingmaAgent: Improving Automated Issue Resolution via Comprehensive Repository Explorationcited by 58
  4. 2025GAM-Agent: Game-Theoretic and Uncertainty-Aware Collaboration for Complex Visual Reasoningcited by 56
  5. 2024AgentFL: Scaling LLM-based Fault Localization to Project-Level Contextcited by 47
  6. 2025Are "Solved Issues" in SWE-bench Really Solved Correctly? An Empirical Studycited by 47
  7. 2024From Code to Correctness: Closing the Last Mile of Code Generation with Hierarchical Debuggingcited by 46
  8. 2025OrcaLoca: An LLM Agent Framework for Software Issue Localizationcited by 43
  9. 2025The Truth Becomes Clearer Through Debate! Multi-Agent Systems with Large Language Models Unmask Fake Newscited by 29
  10. 2023Large Language Models in Fault Localisationcited by 28
  11. 2024Between Lines of Code: Unraveling the Distinct Patterns of Machine and Human Programmerscited by 27
  12. 2025Enhancing repository-level software repair via repository-aware knowledge graphscited by 26
  13. 2024CodeACT: Code Adaptive Compute-efficient Tuning Framework for Code LLMscited by 15
  14. 2024AEGIS: An Agent-based Framework for General Bug Reproduction from Issue Descriptionscited by 15
  15. 2024When Large Language Models Confront Repository-Level Automatic Program Repair: How Well They Done?cited by 12
  16. 2025Putting It All into Context: Simplifying Agents with LCLMscited by 9

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 15; 2026: 37.

52 papers.

Sort:

By area:

  1. 2026SWE-Pruner: Self-Adaptive Context Pruning for Coding Agentscited by 40
  2. 2026Code as Agent Harnesscited by 40
  3. 2025LongCodeZip: Compress Long Context for Code Language Modelscited by 38
  4. 2025SWE-QA: Can Language Models Answer Repository-level Code Questions?cited by 37
  5. 2025Free-MAD: Consensus-Free Multi-Agent Debatecited by 27
  6. 2025Retrieval-Augmented Code Generation: A Survey with Focus on Repository-Level Approachescited by 20
  7. 2026SWE-Explore: Benchmarking How Coding Agents Explore Repositoriescited by 20
  8. 2026In Line with Context: Repository-Level Code Generation via Context Inliningcited by 18
  9. 2026GlimpRouter: Efficient Collaborative Inference by Glimpsing One Token of Thoughtscited by 16
  10. 2025RaCoT: Plug-and-Play Contrastive Example Generation Mechanism for Enhanced LLM Reasoning Reliabilitycited by 14
  11. 2025From Code Foundation Models to Agents and Applications: A Comprehensive Survey and Practical Guide to Code Intelligencecited by 13
  12. 2026Seeing is Coding: On the Effectiveness of Vision Language Models in Code Understandingcited by 13
38 more
  1. 2026SWE-MeM: Learning Adaptive Memory Management for Long-Horizon Coding Agentscited by 12
  2. 2026SWE-QA-Pro: A Representative Benchmark and Scalable Training Recipe for Repository-Level Code Understandingcited by 9
  3. 2026EffiSkill: Agent Skill Based Automated Code Efficiency Optimizationcited by 9
  4. 2026Dockerless: Environment-Free Program Verifier for Coding Agentscited by 8
  5. 2025BOAD: Discovering Hierarchical Software Engineering Agents via Bandit Optimizationcited by 7
  6. 2026AlignCoder: Aligning Retrieval with Target Intent for Repository-Level Code Completioncited by 7
  7. 2025Agentic Software Issue Resolution with Large Language Models: A Surveycited by 6
  8. 2026Test vs Mutant: Adversarial LLM Agents for Robust Unit Test Generationcited by 6
  9. 2026Autonomous Research Agents: A Survey of AI Scientists and the Verification Gapcited by 6
  10. 2026Can Vision-Language Models Handle Long-Context Code? An Empirical Study on Visual Compressioncited by 5
  11. 2026Efficient Failure Management for Multi-Agent Systems with Reasoning Trace Representationcited by 5
  12. 2026SWE-Bench ProMax: Benchmarking Agents on Large-Scale Multilingual Code Refactoringcited by 5
  13. 2026Multi-agent Collaboration with State Managementcited by 4
  14. 2026Code Is More Than Text: Uncertainty Estimation for Code Generationcited by 4
  15. 2025InfCode-C++: Intent-Guided Semantic Retrieval and AST-Structured Search for C++ Issue Resolutioncited by 3
  16. 2025Fed-SE: Federated Self-Evolution for Privacy-Constrained Multi-Environment LLM Agentscited by 3
  17. 2026DRAINCODE: Stealthy Energy Consumption Attacks on Retrieval-Augmented Code Generation via Context Poisoningcited by 3
  18. 2026From Soliloquy to Agora: Memory-Enhanced LLM Agents with Decentralized Debate for Optimization Modelingcited by 3
  19. 2026EquiMem: Calibrating Shared Memory in Multi-Agent Debate via Game-Theoretic Equilibriumcited by 3
  20. 2026LLM Agents Can See Code Repositoriescited by 3
  21. 2025LinkAnchor: An Autonomous LLM-Based Agent for Issue-to-Commit Link Recoverycited by 2
  22. 2025HyperAgent: Leveraging Hypergraphs for Topology Optimization in Multi-Agent Communicationcited by 2
  23. 2025Reliable agent engineering should integrate machine-compatible organizational principlescited by 2
  24. 2026Game-Theoretic Lens on LLM-based Multi-Agent Systemscited by 2
  25. 2026ClassEval-Pro: A Cross-Domain Benchmark for Class-Level Code Generationcited by 2
  26. 2026SHERLOC: Structured Diagnostic Localization for Code Repair Agentscited by 2
  27. 2026SWE-Pruner Pro: The Coder LLM Already Knows What to Prunecited by 2
  28. 2026PhoenixRepair: Rethinking Repair Strategy Exploration in Software Agentscited by 2
  29. 2026SkillForge: Self-Distilling Agents for Project-Specific Issue Resolutioncited by 2
  30. 2025CRAwDAD: Causal Reasoning Augmentation with Dual-Agent Debatecited by 1
  31. 2026Multi-Perspective Agentic Program Repair via Code Property Graphs and Temporal Execution Graphscited by 1
  32. 2026ParaTempo: Efficient Parallel Reasoning via Temporal Confidencecited by 1
  33. 2025From Illusion to Insight: Change-Aware File-Level Software Defect Prediction Using Agentic AIcited by 0
  34. 2026SWE-Spot: Building Small Repo-Experts with Repository-Centric Learningcited by 0
  35. 2026The impact of multi-agent debate protocols on debate quality: a controlled case studycited by 0
  36. 2026Refute-or-Promote: An Adversarial Stage-Gated Multi-Agent Review Methodology for High-Precision LLM-Assisted Defect Discoverycited by 0
  37. 2026When Helping Hurts and How to Fix It: Multi-Agent Debate for Data Cleaningcited by 0
  38. 2026LLVM-Bench: Benchmarking and Advancing Large Language Models for LLVM Compiler Issue Resolutioncited by 0

50 most cited of 52 All 52 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026SWE-Adept: An LLM-Based Agentic Framework for Deep Codebase Analysis and Structured Issue Resolution
  2. 2025SWE-Exp: Experience-Driven Software Issue Resolution
  3. 2026Advances and Frontiers of LLM-based Issue Resolution in Software Engineering: A Comprehensive Survey
  4. 2026Unlocking Model Potentials Through Adaptive Multi-Agent Scaffolding for Efficient Issue Resolution
  5. 2025Trae Agent: An LLM-based Agent for Software Engineering with Test-time Scaling
  6. 2026A Unified Issue Resolution Benchmark for Requirement Clarification, Planning, and Code Generation for Coding Agents
  7. 2026SWE Atlas: Benchmarking Coding Agents Beyond Issue Resolution
  8. 2026BeyondSWE: Can Current Code Agent Survive Beyond Single-Repo Bug Fixing?