Newest papers

Auditing the Hand-Written Inputs of Learned Access Control

Arsalan Jalil Sarghaleh, Mehdi Hazrati · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.31388 · cited by 0 works (OpenAlex) · open-access PDF

Spanish

Learned losks control de acceso se mantiene típicamente desplegado detrás de dos artefactos construidos a mano: un conjunto de denegación dura impuesto por un pequeño escudo de confianza, y un mapa de región que decide cuáles solicitudes ocurren reglas minadas respuesta y cuáles el aprendiz responde. Verificar el escudo o medir el aprendiz no valida ni siquiera. Nosotros auditamos tanto en las cinco políticas públicas de ABAC Lab los modelos ciegos. En primer lugar, formalizamos las restricciones de denegación dura a una sección mientras que la relación de subvención es una regla exhaustiva sobre las relaciones de hasta 794,250 tuples. Cláusulas que configuran fielmente una restricción explícita son sin sonido a tasas de 0.50-0.70 a través de cuatro electores; estos conjuntos de denega

Machine translation (Meta M2M100); the English original follows.

Abstract

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.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Soundness, Set (abstract data type), Ranking (information retrieval), Relation (database), Audit, Computer science

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange