Yang Zhou

Forget painstakingly programming robot interactions – ExoActor uses video generation to hallucinate plausible behaviors, then translates them into robot actions.

Real-time glottis segmentation during Nasotracheal Intubation just got a whole lot faster and more accurate, thanks to a new network that's both lightweight and scale-robust.

DINO, not CLIP, might be the better foundation for open-set 3D object retrieval, especially when paired with dynamic view integration and virtual feature synthesis to avoid overfitting.

Get up to 10% more throughput on your LLM disaggregation workloads just by swapping in this drop-in collective communications library with built-in compression.

Compressing 3D Gaussian Splatting models by iteratively "unfolding" them from a full-resolution version yields surprisingly compact representations without sacrificing rendering quality.

LLMs are far more alike than you think: shared biases and failure modes mean that ensembling them is less effective than you'd hope.

Answering complex questions about 4D scenes just got a whole lot better: PanopticQuery leverages multi-view semantic consensus to transform noisy, view-dependent predictions into globally consistent 4D interpretations.

Explicitly modeling depth in world-action models significantly boosts planning robustness and future prediction quality for autonomous driving.

Diffusion language models can achieve faster convergence and improved accuracy simply by swapping token-choice routing for expert-choice routing, and further benefit from allocating more compute to early denoising steps.

Finally, AI can generate hour-long videos with consistent characters and backgrounds, thanks to a new framework that nails seamless transitions between shots.