2 papers
cs.CV2026
Steering Video Diffusion Transformers with Massive Activations
Xianhang Cheng, Yujian Zheng, Zhenyu Xie +2
In this work, we study the role of Massive Activations (MAs), which are rare, high-magnitude spikes confined to a few fixed hidden dimensions in video diffusion transformers (DiTs)…
cs.CV2025
SOAP: Style-Omniscient Animatable Portraits
Tingting Liao, Yujian Zheng, Adilbek Karmanov +4
Creating animatable 3D avatars from a single image remains challenging due to style limitations (realistic, cartoon, anime) and difficulties in handling accessories or hairstyles.…