2 papers
cs.CV2026
RECAP: Reciprocal Sparse Attention and Hierarchical Caching for Efficient Video Diffusion
Tiange Xu, Kai Liu, Linghe Kong +1
Diffusion transformers generate high-quality videos but remain computationally expensive. Sparse attention and temporal caching reduce computation within and across denoising steps…
cs.CV2026
Fewer Tokens, More Self-Teaching: On-Policy Self-Distillation for Extreme Visual Token Reduction
Junxian Li, Ruixuan Yang, Tianao Zhang +3
Visual token reduction is an effective way to accelerate multimodal large language models (MLLMs), but performance deteriorates rapidly under extremely low token budgets. Existing…