Disen-Forcing: Disentangling Semantic Anchoring from Motion for Autoregressive Video Diffusion

Research Hugging Face Papers Source language: Chinese

论文摘要(作者报告,未经本站复测):Autoregressive video diffusion models (AR-VDMs) enable real-time, long-horizon video generation but degrade rapidly once the inference horizon outruns the training horizon, due to compounding exposure…

Excerpt

论文摘要(作者报告,未经本站复测):Autoregressive video diffusion models (AR-VDMs) enable real-time, long-horizon video generation but degrade rapidly once the inference horizon outruns the training horizon, due to compounding exposure…

Short excerpt from the source — open the original for the full text.

Sources

← Back to the daily brief