Skip to content
Hot eventLive

AliO:提升长期时间序列预测输出对齐

1 reports1 sources3 hr ago updated

Get the story

AI overview

研究者提出 AliO(Align Outputs)方法,用于长期时间序列预测(LTSF)。该方法通过缩小同一时间点在时域与频域上预测输出的差异,提升模型的输出对齐,并引入 TAM(Time Alignment Metric)量化对齐程度。目前报道仅介绍方法思路与评估指标,未披露实验数据或开源信息。

Generated from reports · updated 3 hr ago

Timeline

Follow the coverage from different angles.

Oct 9, 2026
  1. arXiv · Artificial Intelligence
    AliO:长期时间序列预测中的输出对齐至关重要

    研究者提出 AliO(Align Outputs)方法,通过缩小同一时间点在时域与频域上预测输出的差异,提升长期时间序列(LTSF)模型的输出对齐,并引入 TAM(Time Alignment Metric)量化对齐程度。

Heat trend

Current heat 9·Comparable peak 10(Oct 9)·Comparable change over 24 hours –

02.557.510Oct9Oct9Oct9Oct9

The trend compares only the same participants observed continuously; its range may be smaller than the current heat count. Move or click on the chart to inspect hourly heat; use the left and right arrow keys to switch.