TY  - RPRT
TI  - Learning Context-Adaptive Motion Priors for Masked Motion Diffusion Models with Efficient Kinematic Attention Aggregation
AU  - Junkun Jiang
AU  - Jie Chen
AU  - Ho Yin Au
AU  - Jingyu Xiang
PY  - 2026
UR  - https://arxiv.org/abs/2603.07697
ID  - 2603.07697
ER  -