The reviewed record of science sign in
Pith

arxiv: 2402.03317 · v2 · pith:UNYAQM75 · submitted 2024-01-02 · cs.CV · cs.LG

SpecFormer: Guarding Vision Transformer Robustness via Maximum Singular Value Penalization

Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 kernel pith:UNYAQM75record.jsonopen to challenge →

classification cs.CV cs.LG
keywords adversarialattacksspecformervisionvitsboundsmaximummodel
0
0 comments X
read the original abstract

Vision Transformers (ViTs) are increasingly used in computer vision due to their high performance, but their vulnerability to adversarial attacks is a concern. Existing methods lack a solid theoretical basis, focusing mainly on empirical training adjustments. This study introduces SpecFormer, tailored to fortify ViTs against adversarial attacks, with theoretical underpinnings. We establish local Lipschitz bounds for the self-attention layer and propose the Maximum Singular Value Penalization (MSVP) to precisely manage these bounds By incorporating MSVP into ViTs' attention layers, we enhance the model's robustness without compromising training efficiency. SpecFormer, the resulting model, outperforms other state-of-the-art models in defending against adversarial attacks, as proven by experiments on CIFAR and ImageNet datasets. Code is released at https://github.com/microsoft/robustlearn.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. FOCUS: Fused Observation of Channels for Unveiling Spectra

    cs.CV 2025-07 unverdicted novelty 6.0

    FOCUS enables reliable spatial-spectral interpretability for frozen ViTs in hyperspectral imaging with class-specific prompts and a [SINK] token that reduces attention collapse.