Mansi Phute
Scholar

Mansi Phute

Google Scholar ID: D7LxFmgAAAAJ
Georgia Institute of Technology
adversarial machine learningexplainable AI
Citations & Impact
All-time
Citations
337
 
H-index
4
 
i10-index
2
 
Publications
11
 
Co-authors
15
list available
Resume (English only)
Academic Achievements
  • Published 'UNDREAM: Bridging Differentiable Rendering and Photorealistic Simulation for End-to-end Adversarial Attacks', arXiv, 2025.
  • Published 'VISOR++: Transferrable Visual Input based Steering for Output Redirection in Large Vision Language Models', arXiv, 2025.
  • Published 'VISOR: Visual Input based Steering for Output Redirection in Large Vision Language Models', arXiv, 2025.
  • Published 'Semi Truths: A Large-Scale Dataset for Testing Robustness of Image Classifiers', NeurIPS, 2024.
  • Published 'LLM Self Defense: By Self Examination, LLMs Know They Are Being Tricked!', ICLR Tiny Paper, 2024.
  • Some work (e.g., LLM Self Defense) has been deployed at ADP.