Parameter-Efficient Fine-Tuning with Attributed Patch Semantic Graph for Automated Patch Correctness Assessment

2025-05-05

AI Summary

To address the challenge of assessing patch correctness in Automated Program Repair (APR) due to overfitting, this paper proposes the Attributed Patch Semantic Graph (APSG), the first explicit representation that jointly encodes syntactic structure, semantic relationships, and multi-dimensional static code attributes of patches. We further introduce Graph-LoRA, a graph-aware, parameter-efficient fine-tuning method that injects structural priors from APSG into LoRA adapters, enabling lightweight, structure-informed adaptation of large language models (LLMs). Our approach integrates static program analysis, graph neural networks, and LLMs. Evaluated on multiple APR benchmarks, it achieves new state-of-the-art performance: +2.3–6.6% improvement in accuracy and +1.8–6.1% in F1-score. Crucially, it substantially mitigates patch overfitting and enhances discrimination of *intrinsic correctness*, i.e., functional equivalence beyond test-suite satisfaction.

Technology Category

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

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Large language models for search

Abstract

Automated program repair (APR) aims to automatically repair program errors without human intervention, and recent years have witnessed a growing interest on this research topic. While much progress has been made and techniques originating from different disciplines have been proposed, APR techniques generally suffer from the patch overfitting issue, i.e., the generated patches are not genuinely correct despite they pass the employed tests. To alleviate this issue, many research efforts have been devoted for automated patch correctness assessment (APCA). In particular, with the emergence of large language model (LLM) technology, researchers have employed LLM to assess the patch correctness and have obtained the state-of-the-art performance. The literature on APCA has demonstrated the importance of capturing patch semantic and explicitly considering certain code attributes in predicting patch correctness. However, existing LLM-based methods typically treat code as token sequences and ignore the inherent formal structure for code, making it difficult to capture the deep patch semantics. Moreover, these LLM-based methods also do not explicitly account for enough code attributes. To overcome these drawbacks, we in this paper design a novel patch graph representation named attributed patch semantic graph (APSG), which adequately captures the patch semantic and explicitly reflects important patch attributes. To effectively use graph information in APSG, we accordingly propose a new parameter-efficient fine-tuning (PEFT) method of LLMs named Graph-LoRA. Extensive evaluations have been conducted to evaluate our method, and the results show that compared to the state-of-the-art methods, our method improves accuracy and F1 score by 2.3% to 6.6% and 1.8% to 6.1% respectively.

Problem

• Addresses patch overfitting in automated program repair
• Improves semantic capture in patch correctness assessment
• Enhances LLM-based methods with structured patch representation

Innovation

• Attributed Patch Semantic Graph captures patch semantics
• Graph-LoRA enables parameter-efficient LLM fine-tuning
• APSG integrates code attributes and structure