Publication acceptance for the PREDICT project

Many congratulations to the PREDICT team on their manuscript 'Leveraging LLMs for Predictive Insights in Food Policy and Behavioral Intervention' being accepted for publication in the journal Discover Food (accepted July 2025)!

The work of the PREDICT project has been available on the arXiv here (see previous News item), and the paper explores the ability of a fine-tuned large language model (LLM) can to accurately predict the direction of outcomes in empirical studies measuring dietary-based impacts (e.g. food choices, sales, waste) resulting from behavioral interventions and policies, demonstrating approximately 80% accuracy.

Lucia Reisch, Micha Kaiser, Peter Ochieng, and Billy Shi of the YNOT Institute are part of the team working on the PREDICT project, collaborating with Paul Lohmann of the El-Erian Institute at the Cambridge Judge Business School and Cass Sunstein of the Harvard Law School.

Congratulations to the team once again and we look forward to seeing the paper in Discover Food.