Saved in:
| Main Author: | |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2511.18782 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914168644304896 |
|---|---|
| author | Twist, Lukas |
| author_facet | Twist, Lukas |
| contents | Large Language Models (LLMs) often produce code with subtle implementation-level bugs despite strong benchmark performance. These errors are hard for LLMs to spot and can have large behavioural effects; yet when asked to summarise code, LLMs can frequently surface high-level intent and sometimes overlook this low-level noise. Motivated by this, we propose summary-mediated repair, a prompt-only pipeline for program repair that leverages natural-language code summarisation as an explicit intermediate step, extending previous work that has already shown code summarisation to be a useful intermediary for downstream tasks. We evaluate our method across eight production-grade LLMs on two function level benchmarks (HumanEvalPack and MBPP), comparing several summary styles against a direct repair baseline. Error-aware diagnostic summaries consistently yield the largest gains - repairing up to 65% of unseen errors, on average of 5% more than the baseline - though overall improvements are modest and LLM-dependent. Our results position summaries as a cheap, human-interpretable diagnostic artefact that can be integrated into program-repair pipelines rather than a stand-alone fix-all. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_18782 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Summary-Mediated Repair: Can LLMs use code summarisation as a tool for program repair? Twist, Lukas Software Engineering Large Language Models (LLMs) often produce code with subtle implementation-level bugs despite strong benchmark performance. These errors are hard for LLMs to spot and can have large behavioural effects; yet when asked to summarise code, LLMs can frequently surface high-level intent and sometimes overlook this low-level noise. Motivated by this, we propose summary-mediated repair, a prompt-only pipeline for program repair that leverages natural-language code summarisation as an explicit intermediate step, extending previous work that has already shown code summarisation to be a useful intermediary for downstream tasks. We evaluate our method across eight production-grade LLMs on two function level benchmarks (HumanEvalPack and MBPP), comparing several summary styles against a direct repair baseline. Error-aware diagnostic summaries consistently yield the largest gains - repairing up to 65% of unseen errors, on average of 5% more than the baseline - though overall improvements are modest and LLM-dependent. Our results position summaries as a cheap, human-interpretable diagnostic artefact that can be integrated into program-repair pipelines rather than a stand-alone fix-all. |
| title | Summary-Mediated Repair: Can LLMs use code summarisation as a tool for program repair? |
| topic | Software Engineering |
| url | https://arxiv.org/abs/2511.18782 |