Poison-then-Hide: Finetuning-Activated Backdoor Attack on Pretrained Vision Encoders

Abinitha Gourabathina

Abinitha is a PhD student at MIT EECS. She is co-advised by Professors Marzyeh Ghassemi and Collin Stultz. Her research interests lie broadly in trustworthy machine learning. By investigating the latent spaces and distributions of foundational models, she seeks to audit generative models in high-stakes settings like healthcare and improve model performance for safety-driven goals. She completed her B.S.E. in Operations Research and Financial Engineering at Princeton University with minors in Computer Science, Cognitive Science, Linguistics, and Statistics & Machine Learning.