Yong-Jin Liu

A single scene can be represented by ~100K plane primitives (5MB), enabling video world models to maintain geometric consistency across hundreds of frames and large viewpoint changes.

Layer-wise analysis reveals that replay-induced representation drift and optimization dependence are key to understanding catastrophic forgetting in continual learning.

Optimizing function evaluations in flow-based generative models can dramatically enhance image restoration quality without retraining.

By explicitly modeling tooth relationships, TCATSeg achieves state-of-the-art accuracy in 3D dental model segmentation, even in challenging pre-orthodontic cases.

Achieve state-of-the-art semantic scene understanding from sparse views with a feed-forward architecture that generalizes across diverse environments.

Forget static cues: VAGNet grounds 3D object affordances by watching how humans actually use them in videos, significantly improving localization of interaction regions.

By explicitly modeling emotion co-occurrence patterns, MPCL achieves state-of-the-art performance in mixed emotion recognition, outperforming existing methods that neglect the structured correlations among coexisting emotions.