Skip to content
Hot eventLive

LLM引导协方差改进组合半bandit汤普森采样

1 reports1 sources3 hr ago updated

Get the story

AI overview

2026年10月7日，arXiv Statistics Machine Learning（一手来源）发表论文《结构而非信念：组合半bandit中基于LLM导出协方差的关联Thompson采样》，提出一种组合Thompson采样（CTS）改进方法。该方法仅用大语言模型（LLM）对臂做一次划分，经RBF核生成正定协方差矩阵Σ；Beta后验仍只由真实奖励更新，让LLM塑造采样动态而非信念。论文未给出实验结果或具体数值，目前进展限于方法提出阶段。

Generated from reports · updated 3 hr ago

LatestOct 7
新论文提出用LLM导出协方差矩阵塑造CTS采样动态，后验仍由真实奖励更新。

Timeline

Follow the coverage from different angles.

Oct 7, 2026
  1. arXiv · Statistics Machine Learning
    结构而非信念：组合半bandit中基于LLM导出协方差的关联Thompson采样

    论文提出一种组合Thompson采样（CTS）改进方法：仅用LLM对臂做一次划分，经RBF核生成正定协方差矩阵Σ，Beta后验仍只由真实奖励更新，让LLM塑造采样动态而非信念。

Heat trend

Current heat 9·Comparable peak 10（Oct 7）·Comparable change over 24 hours –

02.557.510Oct7Oct7Oct7Oct7

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.