InstaFlow: One Step is Enough for High-Quality Diffusion-Based Text-to-Image Generation
18:28
Diffusion Models (DDPM & DDIM) - Easily explained!
56:16
Flow Matching for Generative Modeling (Paper Explained)
2:00:00
UID Talks - "Shaping The Future of AI Ethics: Equity, Sustainability and Human-centric"
28:39
GaLore: Memory-Efficient LLM Training by Gradient Low-Rank Projection
26:26
InstaFlow: One Step is Enough for High-Quality Diffusion-Based Text-to-Image Generation Explained
17:35
The Reparameterization Trick
22:26
SHViT (CVPR2024): Single-Head Vision Transformer with Memory Efficient Macro Design
1:59:55