Akshay Mehra
Scholar

Akshay Mehra

Google Scholar ID: 5hTbBDMAAAAJ
Tulane University
Adversarial Machine LearningMachine LearningBilevel Optimization
Citations & Impact
All-time
Citations
265
 
H-index
7
 
i10-index
6
 
Publications
20
 
Co-authors
12
list available
Resume (English only)
Academic Achievements
  • Published 'DPCore: Dynamic Prompt Coreset for Continual Test-Time Adaptations' at ICML 2025
  • Published 'Understanding the Transferability of Representations via Task-Relatedness' at NeurIPS 2024
  • Published 'OT-VP: Optimal Transport-guided Visual Prompting for Test-Time Adaptation' at WACV 2025
  • Published 'Test-time Assessment of a Model’s Performance on Unseen Domains via Optimal Transport' at CVPR Workshop 2024
  • Published 'On the Fly Neural Style Smoothing for Risk-Averse Domain Generalization' at WACV 2024
  • Published 'A Spectral View of Randomized Smoothing under Common Corruptions: Benchmarking and Improving Certified Robustness' at ECCV 2022
  • Published 'Understanding the Limits of Unsupervised Domain Adaptation via Data Poisoning' at NeurIPS 2021
  • Published 'How Robust are Randomized Smoothing based Defenses to Data Poisoning?' at CVPR 2021
  • Published 'Penalty Method for Inversion-Free Deep Bilevel Optimization' at ACML 2021
  • Multiple preprints, including works accepted to CVPR 2025 Workshop, NeurIPS 2022 Workshop, etc.