Specificity-Aware Diffusion Steering via Variance-Reduced SM

1. 12:00
   arXiv · Machine Learning Theory

   Specificity-Aware Diffusion Steering via Variance-Reduced Sequential Monte Carlo

   该研究提出一种基于方差缩减序贯蒙特卡洛（SMC）的扩散模型推理期导向方法，通过重叠区域似然比目标抑制负参考分布质量，同时最小化对正分布的扭曲。实验在合成任务、类别对比生成、文本到图像及肽-MHC 结合剂任务上验证，相较负引导基线更有效抑制非期望区域、减小模式偏移并降低 SMC 权重坍缩。代码已公开。

Current heat 9·Comparable peak 10（Oct 2）·Comparable change over 24 hours –

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.