Towards a Human-in-the-Loop Framework for Reliable Patch Evaluation Using an LLM-as-a-Judge

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Shi, Sherry, Wei, Renyao, Tufano, Michele, Cambronero, José, Cheng, Runxiang, Ivančić, Franjo, Rondon, Pat
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908651293245440
author Shi, Sherry
Wei, Renyao
Tufano, Michele
Cambronero, José
Cheng, Runxiang
Ivančić, Franjo
Rondon, Pat
author_facet Shi, Sherry
Wei, Renyao
Tufano, Michele
Cambronero, José
Cheng, Runxiang
Ivančić, Franjo
Rondon, Pat
contents Reliable evaluation is crucial for advancing Automated Program Repair (APR), but prevailing benchmarks rely on execution-based evaluation methods (unit test pass@k), which fail to capture true patch validity. Determining validity can require costly manual annotation. To reduce this cost, we introduce a human-in-the-loop approach to LLM-based patch validity judgment. Inspired by the observation that human judgment is better aligned when using a shared rubric, we first employ an LLM to generate a per-bug rubric, followed by a one-time human review and optional refinement to this rubric, and then employ an LLM to judge patches using the refined rubric. We apply this approach to assign binary validity labels to patches for issues found by Google sanitizer tools. Our results show that this approach yields substantial agreement with human consensus (Cohen's kappa 0.75), high recall (0.94) and high precision (0.80), when considering patches that have unanimous agreement from 3 human raters on the validity labels. On the full dataset including patches where human raters disagree, we find this approach can still be further improved (Cohen's kappa 0.57, recall 0.93, precision 0.65) and identify possible future directions.
format Preprint
id arxiv_https___arxiv_org_abs_2511_10865
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Towards a Human-in-the-Loop Framework for Reliable Patch Evaluation Using an LLM-as-a-Judge
Shi, Sherry
Wei, Renyao
Tufano, Michele
Cambronero, José
Cheng, Runxiang
Ivančić, Franjo
Rondon, Pat
Software Engineering
Reliable evaluation is crucial for advancing Automated Program Repair (APR), but prevailing benchmarks rely on execution-based evaluation methods (unit test pass@k), which fail to capture true patch validity. Determining validity can require costly manual annotation. To reduce this cost, we introduce a human-in-the-loop approach to LLM-based patch validity judgment. Inspired by the observation that human judgment is better aligned when using a shared rubric, we first employ an LLM to generate a per-bug rubric, followed by a one-time human review and optional refinement to this rubric, and then employ an LLM to judge patches using the refined rubric. We apply this approach to assign binary validity labels to patches for issues found by Google sanitizer tools. Our results show that this approach yields substantial agreement with human consensus (Cohen's kappa 0.75), high recall (0.94) and high precision (0.80), when considering patches that have unanimous agreement from 3 human raters on the validity labels. On the full dataset including patches where human raters disagree, we find this approach can still be further improved (Cohen's kappa 0.57, recall 0.93, precision 0.65) and identify possible future directions.
title Towards a Human-in-the-Loop Framework for Reliable Patch Evaluation Using an LLM-as-a-Judge
topic Software Engineering
url https://arxiv.org/abs/2511.10865