Metraipolis › archive › dc_b034aa5f934b

ExploreRepair: Evidence-Specialized Agents for Vulnerability Repair under Information Constraints — arXiv:2610.39899

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.39899
License (as archived)
CC0-1.0
Archived
2026-11-01T02:21:26.247Z
SHA-256 of the archived text
d9769266f91ee440d5e4ad2ef2c2f518f92abf422d717686ebf9370b96d2da8b
Stable URL of this page
https://metraipolis.com/documento/dc_b034aa5f934b

How to cite

ExploreRepair: Evidence-Specialized Agents for Vulnerability Repair under Information Constraints — arXiv:2610.39899 — arXiv (en), https://arxiv.org/abs/2610.39899 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_b034aa5f934b (Metraipolis archive dc_b034aa5f934b, sha256 d9769266f91ee440d5e4ad2ef2c2f518f92abf422d717686ebf9370b96d2da8b)
BibTeX
@misc{dcb034aa5f934b,
  title = {ExploreRepair: Evidence-Specialized Agents for Vulnerability Repair under Information Constraints — arXiv:2610.39899},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.39899},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_b034aa5f934b; sha256 d9769266f91ee440d5e4ad2ef2c2f518f92abf422d717686ebf9370b96d2da8b},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.39899. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Automated vulnerability repair must often operate with incomplete vulnerability information: developers may lack an accurate vulnerability classification or a provided proof-of-vulnerability (PoV) test, yet still need to produce a repair. Without detailed information, a key challenge is to explore multiple plausible interpretations of the vulnerability before committing to a repair. To address this challenge, we propose ExploreRepair, a vulnerability repair framework that reconciles the vulnerability observations and repair implementations from a team of exploration agents. We first collect these observations and repair implementations from three agents, each tasked with exploring a specific dimension of the vulnerability. Then, a Lesson Aggregator and a Patch Aggregator independently synthesize repairs from these observations and concrete implementations before a final Aggregator reconciles their proposals without benchmark feedback. On PatchEval, ExploreRepair achieves repair success rates of 39.6% and 71.6% with GPT-5.6-Luna and DeepSeek-V4.1-Flash, respectively, exceeding EvoRepair, a state-of-the-art vulnerability repair approach, by 16.0 and 25.4 percentage points. On VulnLoc+, it achieves success rates of 51.4% and 62.9%, respectively, exceeding LoopRepair's variant without PoV tests by 20.0 and 17.1 percentage points.

Tema degli esploratori: fiducia digitale e crittografia (elenco arXiv cs.CR, nuovi lavori del giorno).
Categorie arXiv: Software Engineering (cs.SE) ; Cryptography and Security (cs.CR).
Pagina del lavoro: https://arxiv.org/abs/2610.39899 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-01T02:12:13.423Z dalla pagina https://arxiv.org/list/cs.CR/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).