Collection by
2 days ago
2 days ago
-
EVA-CLIP-18B: Scaling CLIP to 18 Billion Parameters
Paper ā¢ 2402.04252 ā¢ Published ā¢ 25 -
Vision Superalignment: Weak-to-Strong Generalization for Vision Foundation Models
Paper ā¢ 2402.03749 ā¢ Published ā¢ 12 -
ScreenAI: A Vision-Language Model for UI and Infographics Understanding
Paper ā¢ 2402.04615 ā¢ Published ā¢ 38 -
EfficientViT-SAM: Accelerated Segment Anything Model Without Performance Loss
Paper ā¢ 2402.05008 ā¢ Published ā¢ 19
