TraceCoder: A Trace-Driven Multi-Agent Framework for Automated Debugging of LLM-Generated Code

2026-02-06

AI Summary

This work addresses the challenge of subtle yet critical errors in code generated by large language models (LLMs), which existing automatic repair methods struggle to fix due to their reliance on coarse-grained pass/fail signals that hinder precise fault localization and often lead to inefficient repair loops. To overcome this, we propose a multi-agent collaborative debugging framework that emulates the human expert workflow of observation, analysis, and repair. Our approach innovatively integrates fine-grained runtime tracing, causal root-cause analysis, a history-based lesson-learning mechanism (HLLM), and a strict rollback strategy to enable accurate and efficient repairs. Experiments demonstrate that our method improves Pass@1 accuracy by up to 34.43% across multiple benchmarks, with the iterative repair module alone yielding a 65.61% relative gain, substantially outperforming existing approaches in both accuracy and cost-effectiveness.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Search and Retrieval-Augmented AI: Large language models for search
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Large Language Models (LLMs) often generate code with subtle but critical bugs, especially for complex tasks. Existing automated repair methods typically rely on superficial pass/fail signals, offering limited visibility into program behavior and hindering precise error localization. In addition, without a way to learn from prior failures, repair processes often fall into repetitive and inefficient cycles. To overcome these challenges, we present TraceCoder, a collaborative multi-agent framework that emulates the observe-analyze-repair process of human experts. The framework first instruments the code with diagnostic probes to capture fine-grained runtime traces, enabling deep insight into its internal execution. It then conducts causal analysis on these traces to accurately identify the root cause of the failure. This process is further enhanced by a novel Historical Lesson Learning Mechanism (HLLM), which distills insights from prior failed repair attempts to inform subsequent correction strategies and prevent recurrence of similar mistakes. To ensure stable convergence, a Rollback Mechanism enforces that each repair iteration constitutes a strict improvement toward the correct solution. Comprehensive experiments across multiple benchmarks show that TraceCoder achieves up to a 34.43\% relative improvement in Pass@1 accuracy over existing advanced baselines. Ablation studies verify the significance of each system component, with the iterative repair process alone contributing a 65.61\% relative gain in accuracy. Furthermore, TraceCoder significantly outperforms leading iterative methods in terms of both accuracy and cost-efficiency.

Problem

• automated debugging
• LLM-generated code
• error localization
• program repair
• runtime traces

Innovation

• trace-driven debugging
• multi-agent framework
• causal analysis
• historical lesson learning
• rollback mechanism