{"slug": "mask-forcing-improving-autoregressive-video-diffusion-distillation-via-dual", "title": "Mask Forcing: Improving Autoregressive Video Diffusion Distillation via Dual-Noise Masking Rollout", "summary": "Researchers propose Mask Forcing, a dual-noise masking rollout technique that improves autoregressive video diffusion distillation by reducing over-saturation artifacts in generated videos. The method targets AR video diffusion models trained via Distribution Matching Distillation (DMD), addressing quality degradation in real-time video generation.", "body_md": "Autoregressive (AR) video diffusion models have shown great potential in real-time video generation. Recent methods distill pretrained bidirectional video diffusion models into causal AR students through Distribution Matching Distillation (DMD), but the generated videos often suffer from over-satura", "url": "https://wpnews.pro/news/mask-forcing-improving-autoregressive-video-diffusion-distillation-via-dual", "canonical_source": "https://aiflash.com/news/116111/", "published_at": "2026-09-09 05:01:14+00:00", "updated_at": "2026-09-09 05:19:54.224771+00:00", "lang": "en", "topics": ["artificial-intelligence", "generative-ai", "machine-learning"], "entities": [], "alternates": {"html": "https://wpnews.pro/news/mask-forcing-improving-autoregressive-video-diffusion-distillation-via-dual", "markdown": "https://wpnews.pro/news/mask-forcing-improving-autoregressive-video-diffusion-distillation-via-dual.md", "text": "https://wpnews.pro/news/mask-forcing-improving-autoregressive-video-diffusion-distillation-via-dual.txt", "jsonld": "https://wpnews.pro/news/mask-forcing-improving-autoregressive-video-diffusion-distillation-via-dual.jsonld"}}