RePaCA: Leveraging reasoning large language models for static automated patch correctness assessment
Marcos Fuster-Pena, David de-Fitero-Dominguez, Antonio García‐Cabot, Eva García‐López · Neurocomputing · 2026
Automated Program Repair (APR) seeks to automatically correct software bugs without requiring human intervention. However, existing tools tend to generate patches that satisfy test cases without fixing the underlying bug. These are known as overfitting patches. To address this issue, Automated Patch Correctness Assessment (APCA) attempts to identify overfitting patches generated by APR tools. This problem can be solved using a static approach, meaning that no additional information is needed beyond the original and fixed code snippets. Current static techniques often struggle with reliability, flexibility and transparency. To address these issues, we introduce RePaCA, a reasoning-based static APCA technique that leverages Large Language Models (LLMs) specialized in thinking tasks. Our model is prompted with both buggy and fixed code snippets and guided to generate a Chain of Thought that analyzes code differences, reasons about how the patch addresses the root cause, and ultimately provides a binary classification: correct or overfitting. To enhance these reasoning capabilities for the APCA task specifically, the LLM is fine-tuned using Reinforcement Learning with the Group Relative Policy Optimization algorithm. When evaluated on a standard Defects4J-derived test, our approach achieves state-of-the-art performance, with 83.1% accuracy and an 84.8% F1-score. Furthermore, our model demonstrates robustness to domain shift when trained on different datasets, outperforming the leading technique. This reasoning capability also provides explicit rationales that can support the inspection of patch assessments. These findings underscore the considerable promise of fine-tuned, reasoning LLMs to advance static APCA by enhancing accuracy and transparency.