Haibin Huang

Forcing moving wrist cameras to predict future pixels actively degrades policy performance—decoupling static scene reconstruction from dynamic wrist latent prediction yields up to a 12.8% boost in contact-rich manipulation with zero inference cost.

Uncertainty DMD is proposed, a simple uncertainty-injection framework that restores stochasticity at two key stages of AR generation: a timestep perturbation for the first chunk to increase first-chunk diversity, and a stochastic cache-writing mechanism for later chunks to preserve uncertainty in autoregressive conditioning.

Unified multimodal models can bootstrap both visual generation and perceptual understanding purely from text prompts by treating their own generative flow trajectories as self-supervised grounding signals.

RefVideo-6M revolutionizes video editing datasets by providing 5 million high-quality editing samples that prioritize visual references over flawed automatic edits.

Uncertainty-guided feedback can dramatically improve the reliability of reward models in visual diffusion, leading to superior optimization and quality outcomes.

CineWeaver breaks the mold by enabling long-form, reference-controllable, and multi-shot video generation without the need for retraining, setting a new standard in cinematic storytelling.

MMOE achieves faster convergence and superior generation quality in diffusion transformers by effectively integrating expert routing strategies, challenging the notion that more parameters always lead to better performance.