Sound Debloating of Redundant Checks in Zero-Knowledge Machine-Learning Circuits
This work presents an automated framework that removes redundant checks while provably preserving soundness in zero-knowledge proof systems, and evaluates circuits spanning MLP, CNN, RNN, and transformer architectures generated by two production frameworks.
Zhan-Tong Xue, Ping-Chuan Ma, Zhaoyu Wang et al.
· 0 citations