Newest papers

Auditor-Assisted Summary-Channel Verification for Hosted LLM Identity Substitution

Ziyi Zhang, Ziyao Wang, Guoheng Sun, Ang Li, Jian Li · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.25982 · cited by 0 works (OpenAlex) · open-access PDF

Korean

호스팅된 합성 LLM 서비스는 점점 더 제품 주장으로 모델 정체성을 노출하지만 외부 사용자와 감사원은 종종 모델 구성이 요청을 제공하는지 검사 할 수 없습니다. 우리는 신뢰할 수있는 감사가 주장 된 정체성을 등록하고 배포 전에 검사 시간 검증 자료를 준비하는 감사가 호스팅되었을 때 모델 내부, 추론 기록, 라우팅 결정, 또는 공급자의 메타데이터에 대한 실행 시간 액세스없이 외부 사용자 정체성과 발표 된 출력만을 사용합니다. SUMMARK은 설치 중에 정체성 특정 단어 운반자를 구축하여 모델 내부, 추론 기록, 경로 결정 또는 공급자 메타데이터에 대한 간단한 어댑터를 훈련합니다. 우리는 배포 시간에 SUMMARK을 제안합니다

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

Abstract

Hosted LLM services increasingly expose model identity as a product claim, but external users and auditors often cannot inspect which model configuration served a request. We study an auditor-assisted regime in which a trusted auditor enrolls claimed identities and prepares verification material before deployment, while audit-time verification uses only the claimed identity and released output, without runtime access to model internals, inference logs, routing decisions, or provider metadata. We propose SUMMARK, a summary-channel watermarking framework that uses user-visible reasoning summaries as a keyed audit surface. During setup, SUMMARK constructs identity-specific vocabulary carriers, trains lightweight summary adapters, and calibrates per-identity thresholds. At deployment time, a keyed statistical test assesses whether the released summary is consistent with the claimed identity and, when needed, attributes the most likely enrolled identity. Across model families, scales, and task domains, SUMMARK reliably detects cross-family and cross-scale substitutions at low false-positive rates while preserving task-facing answer quality. We further study FP16->INT8/INT4 precision-consistency diagnostics and robustness under bounded post-processing, rewriting stress tests, and text-only spoofing, revealing both strong performance and clear failure boundaries. Our code is publicly available at https://github.com/trinitr0toluene/Summark.

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

Topics: Computer science, Identity (music), Rewriting, Computer security, Robustness (evolution), Audit

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