[论文] OPSD-V: On-Policy Self-Distillation for Post-Training Few-Step Autoreg...

论文概要

研究领域: CV 作者: Hongyu Liu, Chun Wang, Feng Gao 发布时间: 2026-07-10 arXiv: 2507.08179

中文摘要

我们提出OPSD-V，一种用于后训练少步自回归（AR）视频扩散模型的在线策略自蒸馏范式。现有少步AR视频生成器可以低延迟生成长视频，但在长自回归展开过程中仍 suffers from 误差累积和运动动力学减弱。OPSD-V在保留原始少步推理路径的同时减少长时程退化。核心思想是在训练期间引入真实长视频数据作为时间上下文，并用其提供密集的轨迹级监督。具体而言，学生遵循精确的推理时间展开，基于自身先前生成的KV缓存生成每个片段。同时，教师在相同的学生访问去噪状态处进行评估，但使用更干净的AR一致时间缓存，其中旧历史可被真实视频上下文替换。这在线策略AR缓存动态下提供了密集的去噪级纠正目标，而不改变采样器、去噪步数或推理时间缓存机制。我们将OPSD-V应用于代表性少步AR视频模型，包括Self-Forcing和LongLive。实验表明在视觉质量、运动动力学和VBenchLong分数上均有持续提升。

原文摘要

We propose OPSD-V, an on-policy self-distillation paradigm for post-training few-step autoregressive (AR) video diffusion models. Existing few-step AR video generators can produce long videos with low latency, but still suffer from error accumulation and weakened motion dynamics during long autoregressive rollout. OPSD-V reduces long-horizon degradation while preserving the original few-step inference path. The key idea is to introduce real long-video data as temporal context during training and use it to provide dense trajectory-level supervision. Specifically, the student follows the exact inference-time rollout, generating each chunk conditioned on its own previously generated KV cache. In parallel, the teacher is evaluated at the same student-visited denoising states, but uses a cleane...

--- *自动采集于 2026-07-11*

#论文 #arXiv #CV #小凯