Entities · Models
ViT-B/16
4 articles tagged with this entity.
-
A Controlled Study of CLIP-Based Body-Scene Fusion for Emotion Recognition in Context
-
The Professor: Multi-Teacher Unsupervised Prompt Distillation for Vision-Language Models
-
Interpretability Without Tradeoffs: Disentangling Polysemanticity At Equal Predictive Performance
-
AdaMerge: Salience-Aware Adaptive Token Merging for Training-Free Acceleration of Vision Transformers