Tushar Nagarajan
Scholar

Tushar Nagarajan

Google Scholar ID: KAKqSwIAAAAJ
FAIR, Meta
Computer visionMachine Learning
Homepage↗Google Scholar↗
Citations & Impact
All-time
Citations
3,848
 
H-index
18
 
i10-index
21
 
Publications
20
 
Co-authors
0
 
Contact
CVOpen ↗GitHubOpen ↗LinkedInOpen ↗
Publications
13 items
RoboJEPA: Scaling Robotic Latent World Models
2026
Cited
0
Cross-Embodiment Robot Foundation World Models with Latent Actions
2026
Cited
0
ThinkJEPA: Empowering Latent World Models with Large Vision-Language Reasoning Model
2026
Cited
0
A Lightweight Library for Energy-Based Joint-Embedding Predictive Architectures
2026
Cited
0
Learning Latent Action World Models In The Wild
arXiv.org · 2026
Cited
4
EgoAdapt: Adaptive Multisensory Distillation and Policy Learning for Efficient Egocentric Perception
2025
Cited
0
PerceptionLM: Open-Access Data and Models for Detailed Visual Understanding
2025
Cited
0
VITED: Video Temporal Evidence Distillation
2025
Cited
0
Resume
Academic Achievements
  • Published multiple papers including PerceptionLM, Switch-a-View, etc., and won 1st place in the EgoSchema Challenge at CVPR 2025.
Research Experience
  • Interned at MALL Lab, IISc, working with Dr. Partha Talukdar.
Education
  • PhD in Computer Science from UT Austin, advised by Dr. Kristen Grauman; B.E. in Computer Science from BITS Goa.
Background
  • Research interests include video understanding and embodied AI. Currently a research scientist at FAIR, Meta.