InfCode: Adversarial Iterative Refinement of Tests and Patches for Reliable Software Issue Resolution

2025-11-19

AI Summary

In real-world warehouse-scale software defect repair, insufficient test coverage and weak validation signals often lead to incorrect patch acceptance. Method: This paper proposes an adversarial multi-agent framework comprising three LLM-based agents—test generation, code generation, and patch selection—that collaboratively and iteratively optimize test cases and patches within a containerized environment to achieve precise fault localization and rigorous validation. Contribution/Results: Its key innovation is a bidirectional adversarial mechanism between testing and code generation, coupled with a failure-driven feedback loop that enhances repair robustness. Evaluated on the SWE-bench Verified benchmark, the framework achieves a 79.4% patch correctness rate—significantly surpassing prior state-of-the-art methods and establishing the new best result. The code and models are publicly released.

Technology Category

• Multiagent Systems: Adversarial Agents
• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Search and Optimization: Adversarial Search

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

Large language models have advanced software engineering automation, yet resolving real-world software issues remains difficult because it requires repository-level reasoning, accurate diagnostics, and strong verification signals. Existing agent-based and pipeline-based methods often rely on insufficient tests, which can lead to patches that satisfy verification but fail to fix the underlying defect. We present InfCode, an adversarial multi-agent framework for automated repository-level issue resolution. InfCode iteratively refines both tests and patches through adversarial interaction between a Test Patch Generator and a Code Patch Generator, while a Selector agent identifies the most reliable fix. The framework runs inside a containerized environment that supports realistic repository inspection, modification, and validation. Experiments on SWE-bench Lite and SWE-bench Verified using models such as DeepSeek-V3 and Claude 4.5 Sonnet show that InfCode consistently outperforms strong baselines. It achieves 79.4% performance on SWE-bench Verified, establishing a new state-of-the-art. We have released InfCode as an open-source project at https://github.com/Tokfinity/InfCode.

Problem

• Automating reliable software issue resolution at repository level
• Addressing insufficient testing that leads to incorrect patches
• Improving verification signals for accurate software defect fixes

Innovation

• Adversarial multi-agent framework refines tests and patches
• Containerized environment enables realistic repository validation
• Iterative refinement between test and code patch generators