Zero-Knowledge Proof Based Verifiable Inference of Models
This addresses the challenge of verifying AI inference for model owners who need to protect intellectual property, representing an incremental advance in verifiable computing for deep learning.
The paper tackles the problem of verifying AI model inference without exposing proprietary model parameters, introducing a zero-knowledge framework that supports various neural network layers and demonstrates practicality with a verifiable version of the DeepSeek model, achieving efficiency and flexibility in real-world workloads.
Recent advances in artificial intelligence (AI), particularly deep learning, have led to widespread adoption across various applications. Yet, a fundamental challenge persists: how can we verify the correctness of AI model inference when model owners cannot (or will not) reveal their parameters? These parameters represent enormous training costs and valuable intellectual property, making transparent verification difficult. In this paper, we introduce a zero-knowledge framework capable of verifying deep learning inference without exposing model internal parameters. Built on recursively composed zero-knowledge proofs and requiring no trusted setup, our framework supports both linear and nonlinear neural network layers, including matrix multiplication, normalization, softmax, and SiLU. Leveraging the Fiat-Shamir heuristic, we obtain a succinct non-interactive argument of knowledge (zkSNARK) with constant-size proofs. To demonstrate the practicality of our approach, we translate the DeepSeek model into a fully SNARK-verifiable version named ZK-DeepSeek and show experimentally that our framework delivers both efficiency and flexibility in real-world AI verification workloads.