# Debug2Fix: Can Interactive Debugging Help Coding Agents Fix More Bugs?

arXiv: https://arxiv.org/abs/2602.18571 · HTML: https://mlsanity.com/paper/2602.18571/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use.
- Builds on 19 of its 27 references that are in the map.
- First posted on arXiv 20 Feb 2026.

## Placed in

- LLM agents and tool use / [Software engineering coding agents](https://mlsanity.com/agents/software-engineering-coding-agents/)

## Builds on (most cited first, 19 of 19)

- [Reflexion: Language Agents with Verbal Reinforcement Learning](https://mlsanity.com/paper/2303.11366/) (2023)
- [Self-Refine: Iterative Refinement with Self-Feedback](https://mlsanity.com/paper/2303.17651/) (2023)
- [SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering](https://mlsanity.com/paper/2405.15793/) (2024)
- [Teaching Large Language Models to Self-Debug](https://mlsanity.com/paper/2304.05128/) (2023)
- [SWE-Bench Pro: Can AI Agents Solve Long-Horizon Software Engineering Tasks?](https://mlsanity.com/paper/2509.16941/) (2025)
- [MapCoder: Multi-Agent Code Generation for Competitive Problem Solving](https://mlsanity.com/paper/2405.11403/) (2024)
- [LLM-Based Multi-Agent Systems for Software Engineering: Literature Review, Vision and the Road Ahead](https://mlsanity.com/paper/2404.04834/) (2024)
- [Debug like a Human: A Large Language Model Debugger via Verifying Runtime Execution Step-by-step](https://mlsanity.com/paper/2402.16906/) (2024)
- [MASAI: Modular Architecture for Software-engineering AI Agents](https://mlsanity.com/paper/2406.11638/) (2024)
- [Are "Solved Issues" in SWE-bench Really Solved Correctly? An Empirical Study](https://mlsanity.com/paper/2503.15223/) (2025)
- [The SWE-Bench Illusion: When State-of-the-Art LLMs Remember Instead of Reason](https://mlsanity.com/paper/2506.12286/) (2025)
- [LiveMCPBench: Can Agents Navigate an Ocean of MCP Tools?](https://mlsanity.com/paper/2508.01780/) (2025)
- [Explainable Automated Debugging via Large Language Model-driven Scientific Debugging](https://mlsanity.com/paper/2304.02195/) (2023)
- [UniDebugger: Hierarchical Multi-Agent Framework for Unified Software Debugging](https://mlsanity.com/paper/2404.17153/) (2024)
- [AutoDev: Automated AI-Driven Development](https://mlsanity.com/paper/2403.08299/) (2024)
- [debug-gym: A Text-Based Environment for Interactive Debugging](https://mlsanity.com/paper/2503.21557/) (2025)
- [Saving SWE-Bench: A Benchmark Mutation Approach for Realistic Agent Evaluation](https://mlsanity.com/paper/2510.08996/) (2025)
- [GitBug-Java: A Reproducible Benchmark of Recent Java Bugs](https://mlsanity.com/paper/2402.02961/) (2024)
- [ChatDBG: Augmenting Debugging with Large Language Models](https://mlsanity.com/paper/2403.16354/) (2024)

## Built on by (most cited first, 1 of 1)

- [Terminus-4B: Can a Smaller Model Replace Frontier LLMs at Agentic Execution Tasks?](https://mlsanity.com/paper/2605.03195/) (2026)

## Abstract (arXiv)

While significant progress has been made in automating various aspects of software development through coding agents, there is still significant room for improvement in their bug fixing capabilities. Debugging and investigation of runtime behavior remains largely a manual, developer-driven process. Popular coding agents typically rely on either static analysis of the code or iterative test-fix cycles, which is akin to trial and error debugging. We posit that there is a wealth of rich runtime information that developers routinely access while debugging code, which agents are currently deprived of due to design limitations. Despite how prevalent debuggers are in modern IDEs and command-line tools, they have surprisingly not made their way into coding agents. In this work, we introduce Debug2Fix, a novel framework that incorporates interactive debugging as a core component of a software engineering agent via a subagent architecture. We incorporate debuggers for Java and Python into our agent framework and evaluate against GitBug-Java and SWE-Bench-Live and achieve >20% improvement in performance compared to the baseline for certain models. Furthermore, using our framework, we're able to make weaker models like GPT-5 and Claude Haiku 4.5 match or exceed the performances of stronger models like Claude Sonnet 4.5, showing that better tool design is often just as important as switching to a more expensive model. Finally, we conduct systematic ablations demonstrating the importance of both the subagent architecture and debugger integration.