1

PromptCAL: Contrastive Affinity Learning via Auxiliary Prompts for Generalized Novel Category Discovery

**Self-Learning**, **CVPR 2023** Novel class discovery through prompting.

Boosting Adversarial Transferability using Dynamic Cues

**Adversarial Machine Learning**, **ICLR 2023** Learning temporal prompts within image models (ViTs) to fool video models.

Stylized Adversarial Defense

**TPAMI 2022**, **Impact Factor [24.31]** Target-aware adversarial training with style, content, and boundary loss.

Guidance Through Surrogate: Towards a Generic Diagnostic Attack

**TNNLS 2022**, **Impact Factor [14.25]** Match and Deceive loss for guidance to break adversarial defenses.

Adversarial Pixel Restoration as a Pretext Task for Transferable Perturbations

**Adversarial Machine Learning**, **BMVC 2022**, **Oral**, **Top 9.5%** Self-supervised adversarial training for transferable attacks.

Self-Distilled Vision Transformer for Domain Generalization

**Self-Learning**, **ACCV 2022**, **Oral**, **Top 14.6%** Self-regularization for Vision Transformer for Domain Generalization.

Self-supervised Video Transformer

**Self-Learning**, **CVPR 2022**, **Oral**, **Top 5.0%** Self-supervised spatiotemporal view matching for video understanding.

On Improving Adversarial Transferability of Vision Transformers

**Adversarial Machine Learning**, **ICLR 2022**, **Spotlight**, **Top 5.0%** Self-ensemble with token/feature refinement for adversarial transferability.

How to Train Vision Transformer on Small-scale Datasets?

**BMVC 2022** Learning self-supervised inductive biases from small data for ViTs.

Intriguing Properties of Vision Transformers

**Visual-Spatial Perception**, **NeurIPS 2021**, **Spotlight**, **Top 3.0%** Analysis of neural behavioral patterns of self-attention within ViTs.