HyperFL: Query-Adaptive Representation Learning for Software Fault Localization

2026-08-03

AI Summary

This work addresses the limitations of existing retrieval-based software defect localization methods, which rely on fixed query representations and struggle to handle the substantial variability in real-world bug reports regarding length, structure, and debugging information. To overcome this, we propose HyperFL, a novel framework that introduces query-adaptive mechanisms to this task for the first time. HyperFL employs a lightweight hypernetwork to dynamically generate query-specific LoRA parameters, thereby personalizing the query encoder while keeping the code encoder fixed to preserve reusability. Integrated within a shared embedding space under the dense retrieval paradigm, this approach significantly enhances the model’s adaptability to diverse bug reports. Experimental results demonstrate that HyperFL achieves a 13.3% relative improvement in function-level MRR@10 and a 16.7% gain in Hit@1 over the current state-of-the-art method, SweRank, on realistic defect localization benchmarks.

Abstract

Software fault localization identifies the code locations responsible for reported issues and is a fundamental step toward automated debugging and program repair. Recent retrieval-based approaches formulate fault localization as a dense retrieval task by learning a shared embedding space between issue reports and source code. However, these methods encode all issue reports using a fixed query representation, despite the substantial diversity of real-world issue reports in length, structure, and debugging information. To address this limitation, we propose HyperFL, a query-adaptive representation learning framework for software fault localization. HyperFL employs a lightweight hypernetwork to generate query-specific LoRA parameters for the query encoder, enabling dynamic query adaptation while keeping the code encoder fixed and reusable. Experiments on a real-world issue localization benchmark demonstrate that HyperFL consistently improves retrieval performance across multiple embedding backbones, achieving up to 13.3% relative improvement in function-level MRR@10 and 16.7% relative improvement in Hit@1 over the state-of-the-art method SweRank. Further analysis shows that HyperFL learns distinct adaptation patterns for different issue characteristics, highlighting the effectiveness of query-adaptive representations for software issue localization.

Problem

• software fault localization
• query representation
• issue report diversity
• dense retrieval
• embedding space

Innovation

• query-adaptive representation
• hypernetwork
• LoRA
• software fault localization
• dense retrieval