All lessons
parallel-decoding
2 free lessons tagged parallel-decoding across AI. Each one is a short sequence of focused steps with narration and a five-question quiz at the end — take them in any order, no signup required.
AI
advancedMaking Diffusion LLMs Actually Fast
Why bidirectional attention breaks the KV cache, how block-wise approximate caching brings it back, and the conditional-independence problem that decides how many tokens you can safely unmask at once.
9 steps·~14 min
AI
advancedDiffusion language models: generating text all at once
How diffusion was rebuilt for discrete text: masked diffusion instead of Gaussian noise, parallel refinement instead of token-by-token decoding, the speed economics that make it attractive, and the honest trade-offs against autoregressive transformers.
8 steps·~12 min