Ziyi Zhang, Ziyao Wang, Guoheng Sun, Ang Li, Jian Li · 2026 · arXiv
DOI: 10.48550/arXiv.2610.25982 · open-access PDF
Layanan LLM yang diselenggarakan semakin mengekspos identitas model sebagai klaim produk, tetapi pengguna eksternal dan auditor sering tidak dapat memeriksa konfigurasi model mana yang melayani permintaan. Kami mempelajari rezim yang dibantu auditor di mana seorang auditor terpercaya mendaftarkan identitas yang diklaim dan mempersiapkan materi verifikasi sebelum implementasi, sementara verifikasi waktu audit hanya menggunakan identitas yang diklaim dan output yang dirilis, tanpa akses runtime ke internal model, log inferensi, keputusan routing, atau metadata penyedia.
Machine translation (Meta M2M100); the English original follows.
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 summarie
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=2610.25982, 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