Course 03
Large Language Models
- Thursday, 3 December 2026
- 12:15 – 13:45
- Online
This session covers large language models: their architecture, text generation, prompting, and the main techniques used to speed up inference.
One thread runs through the whole session: the cost of compute and memory.
On the programme
- 01Definition and architecture
- 02Mixture of experts
- 03Context length, temperature
- 04Sampling strategies
- 05Prompting, in-context learning
- 06Chain of thought
- 07Self-consistency