Orr Zohar


Stanford, California

I am a PhD student in Machine Learning at Stanford University, advised by Serena Yeung and funded by a Knight-Hennessy Scholarship.

I am interested in developing novel machine learning methods that enable the evaluation of actions in video, and their possible applications to surgical procedures. I am particularly interested in open world learning, foundational multimodal models and their potential real-world impact. I then use the methods I develop either for post-surgical AI-driven surgical analytics or AI-assisted surgery.

Before starting my PhD, I was studying towards a Masters of Electrical Engineering at the Technion ECE department, with a focus on Machine Learning and Image Processing. I have a diverse research background, and have been fortunate to collaborate with several talented researcher - spanning from Soft Electronics at LNBD, ultrafast superconductor-based single photon detectors at QUAD, to OCT-based medical imaging at the de la Zerda laboratory.


Mar 13, 2023 Look out for my new paper “LOVM: Language-Only Vision Model Selection”, out on arXiv soon!
Mar 9, 2023 My paper “PROB: Probabilistic Objectness for Open World Object Detection” was accepted to CVPR 2023!
May 1, 2021 Awarded the Knight-Hennessy Scholarship, headed to Stanford University!

selected publications

  1. PROB: Probabilistic Objectness for Open World Object Detection
    Orr Zohar, Kuan-Chieh Wang, and Serena Yeung
  2. Analyzing Surgical Technique in Diverse Open-Surgical Videos with Multi-Task Machine Learning
    Emmett D. Goodman, Krishna K. Patel, Yilun Zhang, and 11 more authors
    TBD (under review) 2023