Unit 4: Alignment and current research
In this unit, you will learn more about the alignment stage of LLM training. You will also see some examples of current research in this and related areas. The unit features both lecture-style reviews of recent developments and videos from research presentations.
Lectures
The lectures start by exploring LLMs alignment. You will then learn about retrieval-augmented generation and current research on how LLMs store facts, and how tokenisation relates to LLM privacy and security. The series concludes by looking at the environmental cost of chatbot technology.
| Section | Title | Video | Slides |
|---|---|---|---|
| 4.1 | Reinforcement learning with human feedback | video | slides |
| 4.2 | Direct preference optimisation | video | slides |
| 4.3 | Retrieval-augmented generation | video | none |
| 4.4 | LLMs for fact completion | video | slides |
| 4.5 | Adversarial tokenization | video | slides |
| 4.6 | Environmental cost of chatbot technology | video | slides |