TeleSparse: Practical Privacy-Preserving Verification of Deep Neural Networks

2025-04-27

AI Summary

This work addresses the fundamental challenge of balancing privacy preservation and computational efficiency in zero-knowledge (ZK) verification of large language model (LLM) inference integrity. We propose the first privacy-preserving ZK-SNARK verification framework tailored for Transformer-based architectures. Methodologically, we introduce a novel “neural sparsification + neural jumping” co-optimization paradigm, integrating structured pruning, activation range compression, lookup-table optimization, and sparse circuit construction to design ZK-friendly models within the Halo2 framework. Experimental evaluation on ViT, ResNet, and MobileNet demonstrates that our approach reduces proof generation time by 46%, decreases memory footprint by 67%, and incurs only ~1% accuracy degradation. These results significantly enhance the practicality and deployability of ZK proofs for large-scale model inference, establishing a new foundation for efficient, privacy-respecting model verification.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Computer Vision: Large Vision Models
• Natural Language Processing: (Large) Language Models

Application Category

• Security and Privacy: Data transparency and provenance
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Large language models for search

Abstract

Verification of the integrity of deep learning inference is crucial for understanding whether a model is being applied correctly. However, such verification typically requires access to model weights and (potentially sensitive or private) training data. So-called Zero-knowledge Succinct Non-Interactive Arguments of Knowledge (ZK-SNARKs) would appear to provide the capability to verify model inference without access to such sensitive data. However, applying ZK-SNARKs to modern neural networks, such as transformers and large vision models, introduces significant computational overhead. We present TeleSparse, a ZK-friendly post-processing mechanisms to produce practical solutions to this problem. TeleSparse tackles two fundamental challenges inherent in applying ZK-SNARKs to modern neural networks: (1) Reducing circuit constraints: Over-parameterized models result in numerous constraints for ZK-SNARK verification, driving up memory and proof generation costs. We address this by applying sparsification to neural network models, enhancing proof efficiency without compromising accuracy or security. (2) Minimizing the size of lookup tables required for non-linear functions, by optimizing activation ranges through neural teleportation, a novel adaptation for narrowing activation functions' range. TeleSparse reduces prover memory usage by 67% and proof generation time by 46% on the same model, with an accuracy trade-off of approximately 1%. We implement our framework using the Halo2 proving system and demonstrate its effectiveness across multiple architectures (Vision-transformer, ResNet, MobileNet) and datasets (ImageNet,CIFAR-10,CIFAR-100). This work opens new directions for ZK-friendly model design, moving toward scalable, resource-efficient verifiable deep learning.

Problem

• Verify deep learning inference without accessing sensitive model weights or data
• Reduce computational overhead of ZK-SNARKs for modern neural networks
• Optimize ZK-SNARK efficiency via sparsification and activation range minimization

Innovation

• ZK-SNARKs with sparsification for efficient verification
• Neural teleportation to minimize lookup table size
• Halo2 proving system for scalable verifiable deep learning