GALA: Multimodal Graph Alignment for Bug Localization in Automated Program Repair

TL;DR

GALA introduces a structured multimodal framework for bug localization in automated program repair, effectively aligning visual UI elements with code components to improve accuracy.

Contribution

It proposes a novel hierarchical structural alignment approach that explicitly models visual and code relationships, surpassing prior methods that rely on plain text conversion.

Findings

GALA achieves state-of-the-art results on the SWE-bench Multimodal benchmark.

The framework effectively captures spatial relationships in UI images for precise bug localization.

Hierarchical alignment significantly improves the mapping between visual observations and code.

Abstract

Large Language Model (LLM)-based Automated Program Repair (APR) has shown strong potential on textual benchmarks, yet struggles in multimodal scenarios where bugs are reported with GUI screenshots. Existing methods typically convert images into plain text, which discards critical spatial relationships and causes a severe disconnect between visual observations and code components, leading localization to degrade into imprecise keyword matching. To bridge this gap, we propose GALA (Graph Alignment for Localization in APR), a framework that shifts multimodal APR from implicit semantic guessing to explicit structural reasoning. GALA operates in four stages: it first constructs an Image UI Graph to capture visual elements and their structural relationships; then performs file-level alignment by cross-referencing this UI graph with repository-level structures (e.g., file references) to locate candidate files; next conducts function-level alignment by reasoning over fine-grained code dependencies (e.g., call graphs) to precisely ground visual elements to corresponding code components; and finally performs patch generation within the grounded code context based on the aligned files and functions. By systematically enforcing both semantic and relational consistency across modalities, GALA establishes a highly accurate visual-to-code mapping. Evaluations on the SWE-bench Multimodal benchmark demonstrate that GALA achieves state-of-the-art performance, highlighting the effectiveness of hierarchical structural alignment.