Daniel LK Yamins

Unified 3D Scene Understanding Through Physical World Modeling

Zero-shot Object Understanding with a Physically Controllable World Model

Self-Supervised Learning of Motion Concepts by Optimizing Counterfactuals

Hypothetical Minds: Scaffolding Theory of Mind for Multi-Agent Tasks with Large Language Models

Taming generative video models for zero-shot optical flow extraction

The BabyView dataset: High-resolution egocentric videos of infants’ and young children’s everyday experiences

Learning Counterfactual Interventions for Self-Supervised Motion Estimation

Modeling Focal Synaptic Degeneration and Neural Plasticity in Ventral Visual Cortex