He Wang

A single spatial token, learned via occupancy prediction on a massive dataset, is surprisingly effective at injecting crucial spatial awareness into vision-language navigation, leading to state-of-the-art performance.

Forget hand-crafted heuristics: this new dynamics-aware policy learns to exploit contact forces in cluttered environments, outperforming traditional methods by 25% in simulation and showing impressive sim-to-real transfer.

Forget expensive data collection: Seed2Scale leverages a small-model/large-model synergy to self-generate high-quality embodied AI training data, starting from just four seed demonstrations.