Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2605.27238 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916050582372352 |
|---|---|
| author | Meng, Qianru Zhang, Xiao Ren, Zhaochun Visser, Joost |
| author_facet | Meng, Qianru Zhang, Xiao Ren, Zhaochun Visser, Joost |
| contents | LLM-based agents have moved automated program repair (APR) from fixed-context patch generation to interactive repository-level repair. However, existing agentic APR systems still struggle to use execution evidence to guide localization, patch generation, and validation. We propose EviACT (Evidence-to-Action), an agentic APR framework that coordinates three evidence-driven guardrails across repair stages. The retrieval scaffold grounds repair context, the compile gate filters invalid edits, and the test-driven gate checks target-test recovery before full regression. Across four benchmarks, EviACT improves resolve rate over the strongest reported comparable baselines by 1.6-6.0 percentage points and shows 70.1-88.6% lower reported per-bug API cost where baseline costs are available. Ablations and diagnostics suggest that these gains are associated with the coordinated evidence-to-action chain, making agentic APR more effective and efficient. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_27238 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | EviACT: An Evidence-to-Action Framework for Agentic Program Repair Meng, Qianru Zhang, Xiao Ren, Zhaochun Visser, Joost Software Engineering LLM-based agents have moved automated program repair (APR) from fixed-context patch generation to interactive repository-level repair. However, existing agentic APR systems still struggle to use execution evidence to guide localization, patch generation, and validation. We propose EviACT (Evidence-to-Action), an agentic APR framework that coordinates three evidence-driven guardrails across repair stages. The retrieval scaffold grounds repair context, the compile gate filters invalid edits, and the test-driven gate checks target-test recovery before full regression. Across four benchmarks, EviACT improves resolve rate over the strongest reported comparable baselines by 1.6-6.0 percentage points and shows 70.1-88.6% lower reported per-bug API cost where baseline costs are available. Ablations and diagnostics suggest that these gains are associated with the coordinated evidence-to-action chain, making agentic APR more effective and efficient. |
| title | EviACT: An Evidence-to-Action Framework for Agentic Program Repair |
| topic | Software Engineering |
| url | https://arxiv.org/abs/2605.27238 |