Synth^2: Boosting Visual-Language Models with Synthetic Captions and Image Embeddings Paper • 2403.07750 • Published Mar 12, 2024 • 24
