发表机构
NVIDIA（英伟达）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对高分辨率视频生成成本高的问题，提出SoL-Refiner单步细化器，通过持续训练、强化学习和蒸馏，实现4K视频单步细化，在2K和4K下均优于现有方法，延迟加速8.91倍。
AI 中文摘要
高分辨率视频生成成本高昂，因为其成本随着时空标记数量的增加而迅速增长。一种实用的替代方案是先生成较低分辨率的视频，然后应用细化器，但传统的多步细化会引入第二个采样瓶颈。我们提出了SoL-Refiner，一种单步视频细化器，通过单次去噪步骤将低分辨率模型输出转换为4K视频。我们的三阶段方案结合了高分辨率持续训练、强化学习（RL）后训练和最终的单步蒸馏。我们引入了Refiner-Bench，这是一个由不同视频生成器的输出构建的视频细化基准，并使用共享输入协议在约2K输出分辨率下比较细化器。在2K分辨率下，单步SoL-Refiner在VBench和UniPercept平均值上优于所有外部细化器，而在3840×2176分辨率下，它在两个指标上都优于三步LTX-2.3细化器。凭借完整的加速堆栈，SoL-Refiner在我们的2K延迟设置中实现了相对于同一基线8.91倍的细化延迟加速。
英文摘要
High-resolution video generation is expensive, as its cost grows rapidly with the number of spatiotemporal tokens. A practical alternative first generates a lower-resolution video and then applies a refiner, but conventional multi-step refinement introduces a second sampling bottleneck. We present SoL-Refiner, a one-step video refiner that transforms low-resolution model outputs into 4K videos with a single denoising step. Our three-stage recipe combines high-resolution continual training, reinforcement learning (RL) post-training, and a final one-step distillation. We introduce Refiner-Bench, a video refinement benchmark constructed from the outputs of different video generators, and use a shared-input protocol to compare refiners at approximately 2K output resolution. At 2K, the one-step SoL-Refiner outperforms all external refiners on the VBench and UniPercept averages, while at $3840\!\times\!2176$ it improves both metrics over the three-step LTX-2.3 Refiner. With the complete acceleration stack, SoL-Refiner achieves an $8.91\times$ speedup in refinement latency over the same baseline in our 2K latency setting.
Comments15 pages
