Leveraging Causal Inference for Explainable Automatic Program Repair

TL;DR

This paper introduces CPR, a causal inference-based method for explainable automatic program repair that generates interpretable causal graphs and improves bug fixing across multiple programming languages.

Contribution

It presents a novel causal inference approach for interpretability in program repair, enhancing both explanation quality and repair performance.

Findings

CPR generates meaningful causal graphs for explanations.

CPR improves bug fixing accuracy across four programming languages.

The method enhances interpretability without sacrificing repair effectiveness.

Abstract

Deep learning models have made significant progress in automatic program repair. However, the black-box nature of these methods has restricted their practical applications. To address this challenge, this paper presents an interpretable approach for program repair based on sequence-to-sequence models with causal inference and our method is called CPR, short for causal program repair. Our CPR can generate explanations in the process of decision making, which consists of groups of causally related input-output tokens. Firstly, our method infers these relations by querying the model with inputs disturbed by data augmentation. Secondly, it generates a graph over tokens from the responses and solves a partitioning problem to select the most relevant components. The experiments on four programming languages (Java, C, Python, and JavaScript) show that CPR can generate causal graphs for reasonable interpretations and boost the performance of bug fixing in automatic program repair.