View PDF HTML (experimental)

Abstract:Large Language Models (LLMs) have substantially advanced Automated Program Repair (APR), yet most existing LLM-based APR methods still rely on trial-and-error to generate patches. Such a strategy explores candidate patches in a weakly structured manner, making it difficult to assess the future potential of search directions and allocate search budget effectively. To address this limitation, we propose Psearch, a search-based patch generation framework for LLM-based APR centered on iterative patch evaluation and refinement. Instead of treating patch generation as repeated independent sampling, Psearch maintains a structured search state over intermediate patches, continuously evaluates the promise of explored search paths, and prioritizes the most promising ones for further refinement. This design enables Psearch to abandon weak directions early and progressively approach correct fixes through long-horizon search. Importantly, Psearch can be integrated with different search algorithms, while our current implementation adopts Monte Carlo Tree Search as one effective instantiation. We evaluate Psearch on five widely used bug and vulnerability benchmarks. Experimental results show that Psearch correctly repairs 201 out of 835 bugs in Defects4J, outperforming all 12 state-of-the-art baselines. Psearch also fixes 27 of 79 vulnerabilities in VUL4J and resolves 164 of 300 issues in SWE-Bench-Lite. Moreover, with a patch size of 16, Psearch reduces monetary cost to roughly 50% of strong baselines while maintaining superior repair effectiveness. These results highlight the effectiveness of Psearch for improving LLM-based APR. The code and results can be found at this https URL

Submission history

From: Haichuan Hu [view email]
[v1] Wed, 2 Jul 2025 15:44:12 UTC (598 KB)
[v2] Sun, 28 Sep 2025 11:48:20 UTC (779 KB)
[v3] Thu, 6 Nov 2025 14:13:45 UTC (787 KB)
[v4] Fri, 14 Nov 2025 08:22:01 UTC (846 KB)
[v5] Mon, 6 Jul 2026 16:39:03 UTC (769 KB)