Sora: The paper you need to read

Papers

• Dall-e 3
  • Improving Image Captioning with Better Use of Captions
    • Dalle-3论文阅读 - nlpcver的文章 - 知乎
• Vq-vae
  • Neural Discrete Representation Learning
• DiT
  • Scalable Diffusion Models with Transformers
• Generating Long Videos of Dynamic Scenes
• VideoGPT: Video Generation using VQ-VAE and Transformer
• Imagen Video: High Definition Video Generation with Diffusion Models
• Align your Latents: High-Resolution Video Synthesis with Latent Diffusion Models
• Photorealistic Video Generation with Diffusion Models
• ViViT: A Video Vision Transformer
• 长上下文训练
  • World Model on Million-Length Video And Language With RingAttention
  • DeepSpeed Ulysses: System Optimizations for Enabling Training of Extreme Long Sequence Transformer Models
• Lumiere: A Space-Time Diffusion Model for Video Generation
• magvit2
  • video vae encoder-decoder
• Patch n’ Pack: NaViT, a Vision Transformer for any Aspect Ratio and Resolution
• Scaling Autoregressive Models for Content-Rich Text-to-Image Generation
• VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training
• VideoPoet: A Large Language Model for Zero-Shot Video Generation
  • 分辨率上不去