AI 中文总结
本研究质疑扩散时间序列预测器迭代采样均有益的观点，提出无标签全局停止准则与伯努利时间步采样器，经八组真实数据集实验验证，可提升预测精度并加快推理速度。
AI 中文摘要
扩散模型为时间序列预测中的不确定性建模提供了自然的方式，但其迭代采样过程常被视为均匀有益的优化过程。本研究通过考察反向扩散过程中预测质量的变化对这一观点提出质疑，发现通常在较高噪声水平下就能恢复通用时间结构，而持续的低噪声优化会引入统计漂移并降低最终预测效果。我们的分析进一步表明，这一现象解释了为何现有方法常倾向于采用相对狭窄的扩散架构和调度设计。基于该观察，我们提出一种无标签全局停止准则，用于检测最优终止点，最终可加快推理速度并提升预测准确性。此外，由于早停会在高噪声区域终止推理，我们提出一种伯努利时间步采样器，将训练集中在该区域同时保留对完整扩散过程的覆盖。在八个真实世界数据集上开展的大量实验表明，与现有方法相比，我们的方法具有更优性能。
英文摘要
Diffusion models offer a natural way to model uncertainty in time series forecasting, yet their iterative sampling process is often treated as a uniformly beneficial refinement procedure. Our study challenges this view by examining how forecast quality evolves throughout reverse diffusion. We find that general temporal structure is often recovered at relatively high noise levels, whereas continued low-noise refinement can introduce statistical drift and degrade the final forecast. Our analysis further suggests that this behavior explains why prior methods often favor relatively narrow diffusion architecture and schedule design. Building on this observation, we propose a label-free global stopping criterion that detects the optimal termination point, eventually speeding up inference and improving predictive accuracy. Additionally, since early stopping terminates inference in high-noise regions, we propose a Bernoulli timestep sampler that concentrates training on this region while preserving coverage of the full diffusion process. Extensive experiments conducted across eight real-world datasets demonstrate the superior performance of our method compared to existing approaches.
