Stanford launches MS&E 319 on efficient LLMs: MoE, quantization, speculative decoding
aminkarbasi · x · 2026-09-21
Amin Karbasi and colleagues (@AnayMehrotra, gvelegkas, Amin Saberi) are teaching Stanford's MS&E 319: Efficient Generative Language Models this fall, tackling how to pick training objectives, architectures and inference algorithms under limited compute budgets.
Topics include efficient pre-training, mixture-of-experts, attention and KV-cache compression, quantization, speculative decoding, LoRA, RLHF, DPO and distillation. Lecture materials and recordings will be posted as the course unfolds — 'just buy more GPUs' won't be the only answer.
More from Companies & People
- Sam Altman to brief UN Security Council in person on AI and security — burny_tech · 2026-09-21
- Runway CEO teases announcement: the question he's been asked for five years gets answered tomorrow — c_valenzuelab · 2026-09-21
- Jensen Huang Slams Altman and Amodei's 'Rogue AI' Narrative as a Bid to Escape Existing Laws — mjdramstead · 2026-09-21
- Foresight's Vision Weekend brings Anthropic's Matt Botvinick, Robin Hanson to SF Nov 13-15 — allisondman · 2026-09-21
- Etched CEO Hires Industry Veterans to Teach AI What Not to Do — Alec_Coughlin · 2026-09-21
- Community Pushes Back on Abuse of tibo, the Codex Limits Point Who Resets Usage — adonis_singh · 2026-09-21