Faith: An Efficient Framework for Transformer Verification on GPUs


Boyuan Feng, Tianqi Tang, Yuke Wang, Zhaodong Chen, Zheng Wang, Shu Yang, Yuan Xie, Yufei Ding, University of California, Santa Barbara


Transformer verification draws increasing attention in machine learning research and industry. It formally verifies the robustness of transformers against adversarial attacks such as exchanging words in a sentence with synonyms. However, the performance of transformer verification is still not satisfactory due to bound-centric computation which is significantly different from standard neural networks. In this paper, we propose \textbf{Faith}, an efficient framework for transformer verification on GPUs. We first propose a semantic-aware computation graph transformation to identify semantic information such as bound computation in transformer verification. We exploit such semantic information to enable efficient kernel fusion at the computation graph level. Second, we propose a verification-specialized kernel crafter to efficiently map transformer verification to modern GPUs. This crafter exploits a set of GPU hardware supports to accelerate verification-specialized operations which are usually memory-intensive. Third, we propose an expert-guided autotuning to incorporate expert knowledge on GPU backends to facilitate large search space exploration. Extensive evaluations show that Faith achieves 2.1 times to 3.4 times (2.6 times on average) speedup over state-of-the-art frameworks.

Open Access Media

USENIX is committed to Open Access to the research presented at our events. Papers and proceedings are freely available to everyone once the event begins. Any video, audio, and/or slides that are posted after the event are also free and open to everyone. Support USENIX and our commitment to Open Access.

@inproceedings {280730,
author = {Boyuan Feng and Tianqi Tang and Yuke Wang and Zhaodong Chen and Zheng Wang and Shu Yang and Yuan Xie and Yufei Ding},
title = {Faith: An Efficient Framework for Transformer Verification on {GPUs}},
booktitle = {2022 USENIX Annual Technical Conference (USENIX ATC 22)},
year = {2022},
isbn = {978-1-939133-29-34},
address = {Carlsbad, CA},
pages = {167--182},
url = {},
publisher = {USENIX Association},
month = jul

Presentation Video