Po-Yao Huang

contrastive learning (4) self-supervised learning (4) masked autoencoder (3) multilingual multimodal (3) zero-shot learning (3) multimodal learning (2) image classification (2) image captioning (2) image retrieval (2) object detection (2) representation learning (2) vision-language model (2) attention diversity (2) zero-shot classification (2) multi-head attention (2) visual-semantic embedding (2) speech synthesis (1) attention mechanism (1) efficient training (1) video recognition (1)