Last Updated 04/2026
Embedded Files
Neha Kalibhat
Neha Kalibhat
I am a researcher at Google DeepMind working in the intersection of AI safety and interpretability. I currently work on black-box interpretability methods to understand and steer model behavior with a specific focus on Responsible AI.
I received my PhD in Computer Science from the University of Maryland, College Park where I was advised by Prof. Soheil Feizi at Center for Machine Learning. My PhD research spanned various deep learning areas including representation learning, multi-modal learning, model interpretability/explainability and generative models. My thesis is titled - Interpreting visual representations and mitigating their failures
During the course of my PhD, I had the privilege of pursuing research internships at Google DeepMind (2024), Google Research (2023) and Meta AI (2021). Before UMD, I worked at Citrix for over 2.5 years on email intelligence, analytics and iOS app development. I received my bachelors degree in Computer Science from PES University in 2017, where I did undergraduate research in the intersection of machine learning and high-performance computing.
nehamk[at]umd[dot]edu
nehamk[at]umd[dot]edu
Research
Research
Interpreting and Controlling Model Behavior via Constitutions for Atomic Concept Edits, AISTATS 2026
Neha Kalibhat, Zi Wang, Prasoon Bajpai, Wenjun Zeng, Drew Proud, Been Kim, Mani Malek
Research at Google DeepMind
Understanding the Effect of using Semantically Meaningful Tokens for Visual Representation Learning, CVPR Workshops 2025
Neha Kalibhat, Priyatham Kattakinda, Arman Zarei, Nikita Seleznev, Samuel Sharpe, Senthil Kumar, Soheil Feizi
Collaboration with CapitalOne Research
Augmentations vs Algorithms: What Works in Self-Supervised Learning, Pre-Print
Warren Morningstar, Alex Bijamov, Chris Duvarney, Luke Friedman, Neha Kalibhat, Luyang Liu, Philip Mansfield, Renan Rojas-Gomez, Karan Singhal, Bradley Green, Sushant Prakash
Collaboration with Google Research
Disentangling the Effects of Data Augmentation and Format Transform in Self- Supervised Learning of Image Representations, NeurIPS - UniReps Workshop 2024
Neha Kalibhat, Warren Morningstar, Alex Bijamov, Luyang Liu, Karan Singhal, Philip Mansfield
Research Internship at Google Research
Identifying Interpretable Subspaces in Image Representations, ICML 2023
Neha Kalibhat, Shweta Bhardwaj, Bayan Bruss, Hamed Firooz, Maziar Sanjabi, Soheil Feizi
Collaboration with Meta AI, CapitalOne Research
Measuring Self-Supervised Representation Quality for Downstream Classification using Discriminative Features, AAAI 2024
Neha Kalibhat, Kanika Narang, Hamed Firooz, Maziar Sanjabi, Soheil Feizi
Research Internship at Meta AI
Adapting Self-Supervised Representations to Multi-Domain Setups, BMVC 2023
Neha Kalibhat, Samuel Sharpe, Jeremy Goodsitt, Bayan Bruss, Soheil Feizi
Collaboration with CapitalOne Research
Understanding Over-parameterization in Generative Adversarial Networks, ICLR 2021
Yogesh Balaji*, Mohammadmahdi Sajedi*, Neha Kalibhat, Mucong Ding, Dominik Stöger, Mahdi Soltanolkotabi, Soheil Feizi
Winning Lottery Tickets in Deep Generative Models, AAAI 2021
Software Troubleshooting using Machine Learning, HiPC 2017
Neha Kalibhat, Shreya Varshini, Chid Kollengode, Dinkar Sitaram, Subramaniam Kalambur
Page updated
Google Sites
Report abuse