Alexander Schwing

A single denoising process now unifies discrete token generation and continuous latent trajectories, enabling diffusion models to jointly reason about token dependencies and global structure.

Chronological text logs routinely confuse distinct physical objects across multi-day video, a failure mode resolved by explicitly clustering grounded visual instances into persistent entity biographies rather than relying on episodic text search.

A surprisingly simple transformer architecture beats specialized models on established human motion prediction benchmarks, unifying pose and trajectory forecasting.

By cleverly repurposing text-to-video diffusion models, VideoSketcher achieves high-quality sequential sketch generation from extremely limited human-drawn sketch data.