Skip to content
Hot eventLive

HEST:用熵训练双曲探针做稀疏激活引导

1 reports1 sources16 hr ago updated

Get the story

AI overview

2026年10月5日,arXiv 平台 Machine Learning Theory 栏目发布一手报道,介绍研究者提出的 Hyperbolic Entropy Steering(HEST)方法。该方法使用仅以模型自身 next-token entropy 为标签的轻量探针,将隐藏状态嵌入 Poincaré ball,并只在熵超过阈值处沿测地线方向进行调整,以实现稀疏激活引导。目前报道仅给出方法框架,未披露实验数据或后续验证结果。

Generated from reports · updated 16 hr ago

Timeline

Follow the coverage from different angles.

Oct 5, 2026
  1. arXiv · Machine Learning Theory
    犹豫也有几何结构:用于稀疏激活引导的熵训练双曲探针

    研究者提出 Hyperbolic Entropy Steering(HEST),用仅以模型自身 next-token entropy 为标签的轻量探针,把隐藏状态嵌入 Poincaré ball,只在熵超阈值处沿测地线方向调整。

Heat trend

Current heat 6·Comparable peak 10(Oct 5)·Comparable change over 24 hours –

02.557.510Oct5Oct5Oct5Oct6

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.