Multi-stage Large Language Model Pipelines Can Outperform GPT-4o in Relevance Assessment.

Julian A. Schnabel, Johanne R. Trippas, Falk Scholer, Danula Hettiachchi

Browse the full WWW paper archive.