Secure Transformer Inference Made Non-interactive
Jiawen Zhang
Network and Distributed System Security (NDSS) Symposium 2025 · Day 1 · Privacy & Cryptography 1
The rapid advancement of transformer models has revolutionized artificial intelligence, powering applications from language translation to content generation and question answering. However, the widespread deployment of these powerful models, particularly in services like OpenAI's ChatGPT, introduces significant privacy concerns as users submit sensitive data through prompts and messages. This talk introduces **Nexus**, a groundbreaking non-interactive protocol for secure transformer inference. It addresses the critical need for privacy-preserving AI while overcoming the substantial computational and communication overheads that have plagued previous secure inference solutions.
AI review
Legitimate cryptographic systems research with a concrete novel contribution: a non-interactive FHE protocol for transformer inference that solves two real bottlenecks (matrix multiplication slot waste and O(N) Argmax) with measurable results. The QuickMax O(log N) improvement over Phoenix's O(N) approach is the kind of specific, verifiable claim that earns a serious look. The open-source CKKS bootstrapping implementation on SEAL 4.0 with CUDA is a genuine ecosystem contribution, not a footnote.