Vision & multimodal

From the CNN that cracked ImageNet to vision Transformers, contrastive image-text models, segmentation, speech, vision-language assistants, and the diffusion lineage behind modern image generation.

• Deep Residual Learning for Image Recognition
• An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
• Segment Anything
• Learning Transferable Visual Models From Natural Language Supervision
• Robust Speech Recognition via Large-Scale Weak Supervision
• Visual Instruction Tuning
• Denoising Diffusion Probabilistic Models
• High-Resolution Image Synthesis with Latent Diffusion Models
• Scalable Diffusion Models with Transformers
• Scaling Rectified Flow Transformers for High-Resolution Image Synthesis