DISK:用于世界模型的动态推理跳过
DISK: Dynamic Inference SKipping for World Models
AI总结:
DISK通过动态推理跳过技术,在无需训练的情况下提升自回归世界模型的视频和轨迹预测效率,同时保持预测精度和视觉质量。
摘要:
我们提出了DISK,一种无需训练的自适应推理方法,用于自回归世界模型。DISK通过双分支控制器和跨模态跳过决策协调两个耦合的扩散转换器,用于视频和自身轨迹,从而在不重新训练的情况下保持运动-外观一致性。我们扩展了高阶潜在差分跳过测试到自回归的链式前向制度,并通过滚动循环传播控制器统计信息以实现长时间 horizon 的稳定性。当在1500个NuPlan和NuScenes样本上使用NVIDIA L40S GPU进行闭环驾驶滚动时,DISK在轨迹扩散上实现了2倍加速,在视频扩散上实现了1.6倍加速,同时保持L2规划误差、视觉质量(FID/FVD)和NAVSIM PDMS分数,证明了在大幅降低成本的情况下,能够实现实用的长horizon视频和轨迹预测。
英文摘要:
We present DISK, a training-free adaptive inference method for autoregressive world models. DISK coordinates two coupled diffusion transformers for video and ego-trajectory via dual-branch controllers with cross-modal skip decisions, preserving motion-appearance consistency without retraining. We extend higher-order latent-difference skip testing to the autoregressive chain-of-forward regime and propagate controller statistics through rollout loops for long-horizon stability. When integrated into closed-loop driving rollouts on 1500 NuPlan and NuScenes samples using an NVIDIA L40S GPU, DISK achieves 2x speedup on trajectory diffusion and 1.6x speedup on video diffusion while maintaining L2 planning error, visual quality (FID/FVD), and NAVSIM PDMS scores, demonstrating practical long-horizon video-and-trajectory prediction at substantially reduced cost.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。