Characterizing Multi-Hunk Patches: Divergence, Proximity, and LLM Repair Challenges

TL;DR

This paper analyzes multi-hunk bug fixes, introduces metrics for divergence and proximity, and reveals that current LLMs struggle with dispersed, complex patches, highlighting the need for divergence-aware repair methods.

Contribution

It introduces HUNK4J, a multi-hunk patch dataset, and proposes divergence and proximity metrics to better understand LLM repair challenges.

Findings

LLMs' success decreases with higher divergence and dispersion.

No LLM succeeds on highly dispersed multi-hunk patches.

The study highlights a gap in LLM capabilities for complex, multi-hunk repairs.

Abstract

Multi-hunk bugs, where fixes span disjoint regions of code, are common in practice, yet remain underrepresented in automated repair. Existing techniques and benchmarks pre-dominantly target single-hunk scenarios, overlooking the added complexity of coordinating semantically related changes across the codebase. In this work, we characterize HUNK4J, a dataset of multi-hunk patches derived from 372 real-world defects. We propose hunk divergence, a metric that quantifies the variation among edits in a patch by capturing lexical, structural, and file-level differences, while incorporating the number of hunks involved. We further define spatial proximity, a classification that models how hunks are spatially distributed across the program hierarchy. Our empirical study spanning six LLMs reveals that model success rates decline with increased divergence and spatial dispersion. Notably, when using the LLM alone, no model succeeds in the most dispersed Fragment class. These findings highlight a critical gap in LLM capabilities and motivate divergence-aware repair strategies.