2 papers
cs.CV2026
Accelerating Text-to-Video Generation with Calibrated Sparse Attention
Shai Yehezkel, Shahar Yadin, Noam Elata +2
Recent diffusion models enable high-quality video generation, but suffer from slow runtimes. The large transformer-based backbones used in these models are bottlenecked by spatiote…
cs.CV2025
TokenVerse: Versatile Multi-concept Personalization in Token Modulation Space
Daniel Garibi, Shahar Yadin, Roni Paiss +6
We present TokenVerse -- a method for multi-concept personalization, leveraging a pre-trained text-to-image diffusion model. Our framework can disentangle complex visual elements a…