[short] Synth: Boosting Visual-Language Models with Synthetic Captions and Image Embeddings

Arxiv Papers - [short] Synth: Boosting Visual-Language Models with Synthetic Captions and Image Embeddings