Le-VLWM
Under Review
Text-Goal Control with Vision-Language World Model
arXiv -code -website -
I work on Pretraining and WM+IDMs for Manipulation. I am advised by Vivek Boominathan, Randall Balestriero, & Ashok Veeraraghavan.
Currently, I am the Founding Engineer at Optica Industries, where I lead ML and Perception. We are backed by Lightspeed, Neo, and Addition.
Previously, I worked on VLAs at Persona AI, Multimodal Intelligence at Sieve, & Energy, Nano Neuro-electronics at the MIT Media Lab, Ajayan Group, & IIT Guwahati. I am also on Google Scholar, GitHub, and LinkedIn.
