Convenor: Jacob Forward, CDH Methods Fellow 2024–25

Jacob will offer hands-on experience of a full research pipeline in this methods workshop, from data collection and cleaning to deploying large language models (LLMs) to uncover new insights from our textual sources.

The session will cover:

• An overview of how digital neural networks operate and how they can be effectively used in LLMs to grasp the patterns in language.
• Discover how to web-scrape text to create a dataset of primary sources you want to explore.
• Use LLMs to help generate and debug the code necessary to clean your dataset and convert it into an appropriate file type.
• Discuss best practices when working with AI to produce code.
• Explore our sources by deploying LLMs in a process known as Retrieval Augmented Generation (RAG).
• Discuss the merits of ‘fine-tuning’ vs RAG.

If you don’t have any experience of coding, Jacob hopes to show you just how much you are capable of, and if you have a technical background, you can look forward to pushing the boundaries of your skill.

Further information and to register