Metraipolis › archive › dc_3e036bed9690

Auditing the Hand-Written Inputs of Learned Access Control — arXiv:2610.31388

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.31388
License (as archived)
CC0-1.0
Archived
2026-10-25T08:35:33.116Z
SHA-256 of the archived text
789970b15182b6a644af62787e3d152ecf610fcdd5bfcf055ef1215a81a56c9e
Stable URL of this page
https://metraipolis.com/documento/dc_3e036bed9690

How to cite

Auditing the Hand-Written Inputs of Learned Access Control — arXiv:2610.31388 — arXiv (en), https://arxiv.org/abs/2610.31388 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3e036bed9690 (Metraipolis archive dc_3e036bed9690, sha256 789970b15182b6a644af62787e3d152ecf610fcdd5bfcf055ef1215a81a56c9e)
BibTeX
@misc{dc3e036bed9690,
  title = {Auditing the Hand-Written Inputs of Learned Access Control — arXiv:2610.31388},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.31388},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3e036bed9690; sha256 789970b15182b6a644af62787e3d152ecf610fcdd5bfcf055ef1215a81a56c9e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.31388. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Learned access control is typically deployed behind two hand-built artefacts: a hard-deny set enforced by a small trusted shield, and a region map that decides which requests mined rules answer and which the learner answers. Verifying the shield or measuring the learner validates neither. We audit both on the five public ABAC Lab policies. First, we formalise hard-deny soundness and check it exhaustively over relations of up to 794,250 tuples. Clauses that faithfully transcribe an explicit restriction are unsound at rates of 0.50-0.70 across four elicitors - an author and three blind language-model elicitors - because prose scopes restrictions to a section while the grant relation is a disjunction over all rules: 97% of counterexamples are granted by other rules. Shipped unchecked, these deny sets would withhold 16-78% of everything the policies grant; certified, they withhold nothing. Vacuity, a second silent failure, also occurs. Second, hybrid routing needs a two-class, abstaining rule set that ABAC miners do not emit, and ranking a mixed-effect candidate pool under a cap is one-class in all 24 cells for the two commonest ranking keys. Third, for additive losses the hybrid-versus-learner effect factorises into covered mass times a rule-learner risk contrast that must be reported on both false-non-grant and false-permit loss: on two real policies, most selection rules give opposite verdicts on the two losses, and the one favourable on both holds for one policy, one learner configuration and a band of noise levels. A released artefact regenerates and checks every reported number.

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).
Pagina del lavoro: https://arxiv.org/abs/2610.31388 — 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 2026-10-25T08:31:13.256Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).