Mask Forcing: Improving Autoregressive Video Diffusion Distillation via Dual-Noise Masking Rollout Paper • 2609.09123 • Published 14 days ago • 55
view article Article Wan 3.0 Ecosystem Signals: What WanSong, Wan-Dancer, and Wan-Streamer Reveal About Alibaba's Next Video Model ResterChed • Jul 29 • 2
Read It Back: Pretrained MLLMs Are Zero-Shot Reward Models for Text-to-Image Generation Paper • 2607.11886 • Published Jul 13 • 60
UnityShots: Memory-Driven Multi-Shot Audio-Video Generation with Boundary-Aware Gating Paper • 2606.21661 • Published Jun 19 • 29
UnityShots: Memory-Driven Multi-Shot Audio-Video Generation with Boundary-Aware Gating Paper • 2606.21661 • Published Jun 19 • 29
UnityShots: Memory-Driven Multi-Shot Audio-Video Generation with Boundary-Aware Gating Paper • 2606.21661 • Published Jun 19 • 29