Skip to content
Hot eventLive

Attention Manifolds：通过编辑B样条曲面引导或阻断语言模型

1 reports1 sources5 hr ago updated

Get the story

AI overview

2026-10-02，arXiv Machine Learning Theory 发表一手论文，提出 Attention Manifolds 方法。该方法在标准 Transformer 注意力之上叠加 learned 2D B-spline 表面 S_d(q_d, k_d)，按 query-key 交互调制每个 value 维度的输出，从而引导或阻断语言模型的行为。论文仅给出方法设计与理论动机，尚未报告具体实验结果或数据集表现。

Generated from reports · updated 4 hr ago

LatestOct 2
论文于 arXiv Machine Learning Theory 上线，提出在注意力层叠加 learned 2D B-spline 表面以调制 value 输出。

Timeline

Follow the coverage from different angles.

Oct 2, 2026
  1. arXiv · Machine Learning Theory
    Attention Manifolds：通过编辑学习到的 B-Spline 表面引导或阻断语言模型

    论文提出 attention manifolds：在标准 Transformer 注意力之上叠加 learned 2D B-spline 表面 S_d(q_d, k_d)，按 query-key 交互调制每个 value 维度的输出。

Heat trend

Current heat 9·Comparable peak 9（Oct 2）·Comparable change over 24 hours –

02.557.510Oct2Oct2Oct2Oct2

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.