Anubhav Jain

I am a PhD Candidate in the Computer Science Dept at New York University, where I work on computer vision and machine learning. I am supervised by Dr. Nasir Memon and Dr. Julian Togelius.

I completed my undergrad from IIIT-D, where I wrote my undergraduate thesis with Dr. Mayank Vatsa and Dr. Richa Singh on Image Forensics.

Previously, I was research trainee at Idiap Research Institute in Switzerland, supervised by Dr. Sebastien Marcel. I worked on generalizability in deepfake detection during my time there. Prior to that, I was an Artifical Intelligence Engineer at Digital Product School by UnternehmerTUM in Munich, Germany, where I was supervised by Dr. Afsaneh Asaei. I have also worked briefly at Telstra and Pixxel (space tech start-up).

Email  /  CV  /  Google Scholar  /  Github  /  Linkedin

profile photo
Research

I am exploring problems in Responsible AI where I am trying to make generative models more secure, fair and privacy preserving using little to no training data. More generally I am interested in Multi-modal generative models and controllable image generation.

News

Feb 25: [CVPR 2025] Paper on Memorization in Diffusion models got accepted at CVPR 2025!!

Jan 25: I received the Nokia Bell Labs Scholarship.

Dec 24: Paper on Concept Erasure in Diffusion Models in now up on ArXiv.

Nov 24: Paper on Memorization in Diffusion Models in now up on ArXiv.

May 24: Started my internship at Sony AI on Deep Generative Models in the MFMT team. I'll be working on safe image generation from diffusion models.

May 24: Paper on fair image generation from StyleGAN2 accepted at IEEE Transactions on Biometrics, Behavior, and Identity Science (T-BIOM).

Nov 23: Paper accepted at WACV Workshops.

Sept 23: Presented paper at IJCB 2023 in Ljubljana, Slovenia.

June 23: Paper accepted at IJCB 2023. Rated amongst best reviewed papers and got invited to submit extension to TBIOM.

TraSCE: Trajectory Steering for Concept Erasure
Anubhav Jain, Yuya Kobayashi , Takashi Shibuya , Yuhta Takida , Nasir Memon , Julian Togelius, Yuki Mitsufuji
preprint , 2024
arXiv / Codebase

Classifier-Free Guidance inside the Attraction Basin May Cause Memorization
Anubhav Jain, Yuya Kobayashi , Takashi Shibuya , Yuhta Takida , Nasir Memon , Julian Togelius, Yuki Mitsufuji
CVPR , 2025
arXiv / Codebase

Zero-shot racially balanced dataset generation using an existing biased StyleGAN2
Anubhav Jain, Nasir Memon , Julian Togelius
IEEE/IAPR IJCB, 2023
arXiv / Codebase

Fair GANs through model rebalancing using synthetic data
Anubhav Jain, Nasir Memon , Julian Togelius
preprint, 2023
arXiv

A DataLess FaceDwap Detection Approach Using Synthetic Images
Anubhav Jain, Nasir Memon , Julian Togelius
IEEE/IAPR IJCB, 2022
Paper / Codebase

Dictionary Attacks on Speaker Verification
Mirko Marras , Pawel Korus, Anubhav Jain, Nasir Memon , Julian Togelius
IEEE TIFS, 2022
Paper / Codebase

Custom Attribution Loss for Improving Generalization and Interpretability of Deepfake Detection
Pavel Korshunov , Anubhav Jain, Sebastien Marcel,
ICASSP, 2022
Paper / Codebase

Improving generalization of deepfake detection by training for attribution
Anubhav Jain, Pavel Korshunov , Sebastien Marcel,
MMSP, 2021
Paper / Codebase

Voice analysis to differentiate the dopaminergic response in people with Parkinson's disease
Anubhav Jain, Kian Abedinpour, Ozgur Polat, Mine Melodi Çalışkan, Afsaneh Asaei, Franz MJ Pfister, Urban M Fietzek, Milos Cernak
Frontiers in Human Neuroscience, 2021
Paper

Parameter Estimation and Imaging of SAR Ground Moving Target using AGFS
Peeyush Sahay, Anubhav Jain, P Radhakrishna, Vikram M Gadre
IEEE International Conference on Electronics, Computing and Communication Technologies (CONECCT) , 2021
Paper

Benchmark Dataset for Timetable optimization of Bus Routes in the City of New Delhi
Avdesh Kumar*, Saumya Balodi*, Anubhav Jain*, Pravesh Biyani
IEEE 23rd International Conference on Intelligent Transportation Systems (ITSC), 2020
Paper

Detecting GANs and retouching based digital alterations via DAD-HCNN
Anubhav Jain*, Puspita Majumdar*, Richa Singh, Mayank Vatsa
CVPR Workshops, 2020
Paper

On detecting GANs and retouching based synthetic alterations
Anubhav Jain, Richa Singh, Mayank Vatsa
IEEE 9th international conference on biometrics theory, applications and systems (BTAS) , 2018
Paper


Website source code was borrowed from Jon Barron.