提出新采样方法，让视频生成更快更高效。
Draft-and-Target Sampling for Video Generation Policy
- 用两步互补去噪路径，快速生成再精准验证。
- 在三个基准上最快提升2.1倍速度，成功率几乎不变。
- 适合需要实时视频生成的机器人任务场景。
视频生成模型已被用作机器人策略，根据任务描述和观测预测未来状态。以往工作忽视其高计算成本和长推理时间。为此，我们提出无需训练的新型扩散推理范式——Draft-and-Target Sampling，用于视频生成策略。通过在单一模型中引入自对弈去噪机制，利用两条互补的去噪轨迹：草稿采样以大步快速生成全局轨迹，目标采样以小步精确验证。为进一步加速生成，引入令牌分块与渐进接受策略，减少冗余计算。在三个基准上的实验表明，该方法可实现最高2.1倍的速度提升，且对当前最先进方法的效率改进显著，成功率损失极小。代码已开源。
原文摘要 · Abstract (English)
Video generation models have been used as a robot policy to predict the future states of executing a task conditioned on task description and observation. Previous works ignore their high computational cost and long inference time. To address this challenge, we propose Draft-and-Target Sampling, a novel diffusion inference paradigm for video generation policy that is training-free and can improve inference efficiency. We introduce a self-play denoising approach by utilizing two complementary denoising trajectories in a single model, draft sampling takes large steps to generate a global trajectory in a fast manner and target sampling takes small steps to verify it. To further speedup generation, we introduce token chunking and progressive acceptance strategy to reduce redundant computation. Experiments on three benchmarks show that our method can achieve up to 2.1x speedup and improve the efficiency of current state-of-the-art methods with minimal compromise to the success rate. Our code is available.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。