BumbleBee: Secure Two-party Inference Framework for Large Transformers

Wen-jie Lu

Network and Distributed System Security (NDSS) Symposium 2025 · Day 1 · Privacy & Cryptography 1

This article delves into "BumbleBee," a novel secure two-party inference framework designed for large transformer models, presented by Wen-jie Lu from Zjanu at the NDSS Symposium. The talk addresses a critical challenge in the era of pervasive AI services: how to leverage powerful machine learning models without compromising the privacy of sensitive input data. As individuals and organizations increasingly rely on cloud-based AI, the necessity to transmit confidential information to remote servers raises significant concerns about potential data exposure, misuse, or mismanagement.

AI review

Solid applied-crypto research on a real and underserved problem — private inference for large transformers without trusted hardware. The contributions are genuine (IR-level graph rewriting, numerically stable approximations, 100x communication reduction) but the work sits comfortably within an established research lineage rather than breaking new ground. Right venue, right lane, competent execution.

Watch on YouTube