Skip to content
AI IntelligenceMay 18, 2026AI Intelligence
Article

Tsinghua and Alibaba introduce ViT³, a Vision Transformer architecture achieving linear computational complexity for visual tasks.

This enables practical high-resolution image understanding on edge devices – a major step for real-world AI deployment.

Data Cube AI EditorialSource: Pandaily
01

Source Brief

Tsinghua and Alibaba introduce ViT³, a Vision Transformer architecture achieving linear computational complexity for visual tasks. This enables practical high-resolution image understanding on edge devices – a major step for real-world AI deployment.

02