研究人员推出了一种名为“Next Forcing”的新型多块预测框架，旨在增强自回归视频生成。该方法通过提供关于未来动态的明确信号来解决当前模型的局限性，从而实现更快的训练收敛和更高的准确性，尤其是在高帧率下。该框架还加速了推理，并展示了在生成的视频中更好地遵守物理定律。 AI
影响 加速自回归视频模型的训练和推理，可能实现更复杂、更逼真的视频生成。
排序理由 该集群包含一篇详细介绍视频生成新框架的研究论文。
在 Hugging Face Daily Papers 阅读 →
AI 生成摘要 · Google Gemini · 来自 4 个来源。 我们如何撰写摘要 →
Autoregressive video generation has emerged as a powerful paradigm for World Action Models (WAMs). However, existing approaches suffer from slow training convergence and limited converged accuracy, particularly at high frame rates, as the training supervision is confined to the c…
Next Forcing introduces a multi-chunk prediction framework that accelerates training and inference for autoregressive video generation while improving accuracy and physical law adherence.
arXiv:2606.11187v1 Announce Type: new Abstract: Autoregressive video generation has emerged as a powerful paradigm for World Action Models (WAMs). However, existing approaches suffer from slow training convergence and limited converged accuracy, particularly at high frame rates, …
Autoregressive video generation has emerged as a powerful paradigm for World Action Models (WAMs). However, existing approaches suffer from slow training convergence and limited converged accuracy, particularly at high frame rates, as the training supervision is confined to the c…