Metraipolis › archive › dc_8dd6cfc849f2

P2Patch: Exploit-Guided Vulnerability Repair — arXiv:2610.20245

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.20245
License (as archived)
CC0-1.0
Archived
2026-10-15T20:02:59.184Z
SHA-256 of the archived text
ce300fbea788f9f4a8af417403fdf3595c52ac2202dda7df4c42536f7d83ea05
Stable URL of this page
https://metraipolis.com/documento/dc_8dd6cfc849f2

How to cite

P2Patch: Exploit-Guided Vulnerability Repair — arXiv:2610.20245 — arXiv (en), https://arxiv.org/abs/2610.20245 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_8dd6cfc849f2 (Metraipolis archive dc_8dd6cfc849f2, sha256 ce300fbea788f9f4a8af417403fdf3595c52ac2202dda7df4c42536f7d83ea05)
BibTeX
@misc{dc8dd6cfc849f2,
  title = {P2Patch: Exploit-Guided Vulnerability Repair — arXiv:2610.20245},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.20245},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_8dd6cfc849f2; sha256 ce300fbea788f9f4a8af417403fdf3595c52ac2202dda7df4c42536f7d83ea05},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.20245. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language models are increasingly deployed as autonomous agents for automated program repair, patching software vulnerabilities through multi-step reasoning and tool use. Yet these systems typically validate a patch against the same evidence used to construct it -- a static alert, sanitizer trace, published exploit, or test suite. A patch that blocks that one piece of evidence can therefore look successful while remaining vulnerable to the same vulnerability with some variation. We present P2Patch, a generative adversarial repair system that treats each patch as a new attack target. Starting from a normalized vulnerability finding, an exploitation agent constructs a reproducing proof of vulnerability (PoV), a patching agent repairs the program, and the exploitation agent then attacks the patched program for a distinct same-vulnerability exploit. Every successful exploit becomes a permanent repair requirement, and the final patch must block all PoVs accumulated during the patching process, with construction-time PoVs kept strictly separate from held-out evaluation oracles. Across 101 C/C++ and Java CVE subjects, this adversarial loop surfaces a distinct confirmed exploit in 41% of runs after the patch has already blocked its original construction-time PoV. Against single-pass repair from the same finding using Claude Code, P2Patch raises held-out fix-closed coverage from 78.1% to 84.7% and beyond-upstream coverage from 42.1% to 62.5%. Under the same held-out oracles, it also certifies more attack paths as blocked than recent state-of-the-art agentic repair systems on overlapping subjects. These results show that adversarial testing and independent security oracles expose repair gaps that one-shot validation misses.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Cryptography and Security (cs.CR) ; Artificial Intelligence (cs.AI); Software Engineering (cs.SE).
Pagina del lavoro: https://arxiv.org/abs/2610.20245 — 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 2026-10-15T19:35:48.414Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).