Zhiping Zhang, Franklin Mingzhe Li, Zihao Huang, Tianshi Li · 2026 · arXiv
DOI: 10.48550/arXiv.2610.27548 · open-access PDF
LLM 에이전트가 사용자를 대신하여 행동할 때, 사용자가 허가하는 권한 범위 내에서 행동할 것으로 예상됩니다. 기존 에이전트 평가는 주로 작업 완료 또는 해로운 결과를 측정하고, 따라서 전통적인 보안 또는 개인 정보 보호 손해를 유발하지 않고도 이 임무를 위반하는 행동을 무시할 수 있습니다. we introduce AuthorizationBench, a benchmark of authorization alignment based on documented real-world cases. A scalable construction pipeline produces 286 executable tasks spanning situational triggers, the presence or absence of direct harmful consequences, and ten application.
Machine translation (Meta M2M100); the English original follows.
When an LLM agent acts on behalf of a user, it is expected to act within the authorization scope the user grants. Existing agent evaluations largely measure task completion or harmful consequences and can therefore overlook actions that violate this delegation without producing conventional safety or privacy harms. We introduce AuthorizationBench, a benchmark of authorization alignment grounded in documented real-world cases. A scalable construction pipeline produces 286 executable tasks spanning situational triggers, the presence or absence of direct harmful consequences, and ten application
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.27548, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange