Synthetic data
Can LLMs generate respondent panels that pass standard quality checks? When they do not, where do they fail, and can we measure it before we use them?
Nuffield College Est. 2023
Oxford LLMs · Research · Teaching · Community
Large Language Models for Social Science.
We run an annual summer school at Nuffield College, develop an open-source course on the fundamentals of LLMs for social scientists, and work on methods research. Our faculty spans academic research and the ML industry.
Oxford LLMs 2026, Nuffield College
Research
Methods for LLMs in social science.
We develop methods for applying language models to social science research. Code and datasets are open source on GitHub and Hugging Face.
Opinion prediction
How well do open-weights models recover real survey responses across languages, age cohorts, and political contexts, and what drives the gaps?
Evaluation
Building eval sets that survive contamination, treating annotator disagreement as a measurement signal rather than noise.
Annual intensive
Five days at Nuffield.
Early-career researchers. Lectures, labs, collaborative research. Four editions run since 2023.
2026 Cohort
Six teams. 5 days. Nuffield College.
Lectures from practitioners on agents and evaluation, continual learning, compression, and inference economics, wrapped around a week-long prediction challenge.
Past schools
Four years of cohorts.
Open courseware
Language models for social science.
A five-module course mirroring the summer-school syllabus. Taught by invitation at Oxford, ESSCA Paris, and the EUI. All materials on GitHub.
Supported by 
Past sponsors 