RadZero: Similarity-Based Cross-Attention for Explainable Vision-Language Alignment in Radiology with Zero-Shot Multi-Task Capability Paper • 2504.07416 • Published Apr 10, 2025 • 3
AIMv2 Collection A collection of AIMv2 vision encoders that supports a number of resolutions, native resolution, and a distilled checkpoint. • 16 items • Updated 19 days ago • 82