Jitendra Malik

Rodrigues Network for Learning Robot Actions

D-REX: Differentiable Real-to-Sim-to-Real Engine for Learning Dexterous Grasping

Learning to Grasp Anything By Playing with Random Toys

FoundationMotion: Auto-labeling and Reasoning about Spatial Movement in Videos

ViTacFormer: Learning Cross-Modal Representation for Visuo-Tactile Dexterous Manipulation

DexGarmentLab: Dexterous Garment Manipulation Environment with Generalizable Policy

Whole-Body Conditioned Egocentric Video Prediction

Gaussian Masked Autoencoders

AutoEval Done Right: Using Synthetic Data for Model Evaluation

OTTER: A Vision-Language-Action Model with Text-Aware Visual Feature Extraction

Wolf: Accurate Video Captioning with a World Summarization Framework

Token to Token Learning From Videos

Scaling Diffusion Models for Downstream Prediction

Early Fusion Helps Vision Language Action Models Generalize Better

Poly-Autoregressive Modeling for Interacting Entities