About me

My research focuses on detecting and understanding unexpected phenomena in machine learning models and scientific data.

I am now a post-doc at New York University, working with Prof. Chinmay Hegde. Prior to that I received my PhD at the Hebrew University, supervised by Prof. Yedid Hoshen.

You can also check my personal site with some nice riddles and more.

News

• Our work When Are Concepts Erased From Diffusion Models? will be presented in NeurIPS 2025.

• Our work SEAL: Semantic Aware Image Watermarking will be presented in ICCV 2025.

• I received the Blavatnik Prize for Outstanding Israeli Doctoral Students in Computer Science.

• Our workshop DataWorld: Unifying Data Curation Frameworks Across Domains will take place at ICML 2025. We are accepting submissions until May 21, 2025!

• Our work Hidden in the Noise: Two-Stage Robust Watermarking for Images will be presented in ICLR 2025.

• Three papers I co-authored have been accepted to NeurIPS 2024 (on Tabular Data - Main Track, Data Curation - D & B, and Jailbreaking - D & B Spotlight)

• Our workshop Red Teaming GenAI: What Can We Learn from Adversaries will take place at NeurIPS 2024. We are accepting submissions until September 14, 2024!

• My team together with Yuval Lemberg, Hestia, won 1st place (Defense track) and 3st place (Attack track) in the Large Language Model Capture-the-Flag Competition SaTML 2024.

• Our work Circumventing concept erasure methods for text-to-image generative models will be presented in ICLR 2024.