**Training Vision Transformers to Focus: Emergence of Spatial Attention Head Spe

**Training Vision Transformers to Focus: Emergence of Spatial Attention Head Specialization**

> Self-supervised learning in Vision Transformer models, such as DINO, enables robust visual perception without requiring labeled data by leveraging large-scale image datasets. However, these models lack a focusing mechanism during training, resulting in perceptual representations that are often ambiguous or overly similar. In this work, we introduce two visual focusing loss functions designed to establish corresponden...

**来源信息**
- **来源**：HuggingFace Daily Papers（社区热门论文）
- **分类**：ai-models
- **发布时间**：2026-11-10 08:00（北京时间）
- **原文**：[打开原文](https://huggingface.co/papers/2611.11890)