Hot eventLive
ViTok:AM-RADIO 式多教师蒸馏结合 PHI-S 与 MIM
1 reports1 sources16 hr ago updated
Get the story
AI overview
ViTok 提出一种 AM-RADIO 式学生模型方案,由 SigLIP2 与 DINOv3-L 蒸馏而来,旨在兼顾全局识别与密集语义。该方案采用 CLS/patch 分离适配头、非对称 cosine/MSE 损失、DINOv3-L 初始化、教师重加权、掩码图像建模(MIM)与 PHI-S 特征平衡。目前报道仅给出方法构成,未披露实验数据、基准结果或代码开源情况。
Generated from reports · updated 16 hr ago
Timeline
Follow the coverage from different angles.
Oct 5, 2026
- arXiv · Computer VisionViTok:用 PHI-S 与掩码图像建模改进 AM-RADIO 式多教师蒸馏的密集语义
ViTok 将 AM-RADIO 式学生模型(由 SigLIP2 与 DINOv3-L 蒸馏)整合为兼顾全局识别与密集语义的多教师蒸馏方案,采用 CLS/patch 分离适配头、非对称 cosine/MSE 损失、DINOv3-L 初始化、教师重加权、MIM 与 PHI-S 特征平衡。
Heat trend
Current heat 7·Comparable peak 10(Oct 5)·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.