wisconsin
Practice Questions: Attention Variants, SSMs & Multimodal Models

Practice Questions: Attention Variants, SSMs & Multimodal Models

Content locked

The catalog is public. Reading this content requires approved access.

Sign in to read
1. Quadratic Complexity of Attention [Attention Complexity]
2. Sparse Attention Patterns [Sliding Window / Global Tokens]
3. FlashAttention: Memory Hierarchy [FlashAttention]
4. FlashAttention: Tiling Computation [FlashAttention / Numerical]
5. State-Space Models: Dual View [S4 / SSMs]
6. Mamba vs. S4 [Mamba / Selective State Spaces]
7. Vision Transformer (ViT) [ViT]
8. CLIP: Contrastive Training and Zero-Shot Classification [CLIP]
9. Early vs. Late Fusion [Multimodal Fusion]
10. Flamingo Architecture [Flamingo]
11. LLaVA and Whisper [LLaVA / Whisper]
12. Grounding Problem and SayCan [Grounding / Robotics]