Source · Tier A · Peer-reviewedSource · NanoZK: Privacy-Preserving Verifiable Inference for Large Language Models via Layerwise Zero-Knowledge Proofs
NanoZK: Privacy-Preserving Verifiable Inference for Large Language Models via Layerwise Zero-Knowledge Proofs
Z. Wang. 2026. International Conference on Information and Communications Security (ICICS 2026).
| Original | https://arxiv.org/abs/2603.18046 |
|---|---|
| DOI | 10.48550/arXiv.2603.18046 |
| arXiv | 2603.18046 |
| Version | Current arXiv HTML read on 2026-09-25; its abstract matches v2, which states acceptance at the 28th ICICS (Springer LNCS, Fukui, 27–30 October 2026) and is an extended version with appendices not in the proceedings. v1 (March 2026) reports different timings. The abstract gives 3.5–3.7 KB per sub-circuit proof, while §3.1 and §6.1 give 3.2–3.7 KB. |
| Accessed | 2026-09-25 |
| Imported from | hodgkins-ai-verification-papers@c71e59ff0e8e |
| Note | Listed under "Zero-knowledge proofs" in the Hodgkins bibliography (CC BY 4.0). |