
Jiwoo Hong
Co-author of ORPO: Monolithic Preference Optimization without Reference Model.
Best podcasts with Jiwoo Hong
Ranked by the Snipd community

Jun 13, 2024 • 36min
Fine-tuning and Preference Alignment in a Single Streamlined Process
Jiwoo Hong and Noah Lee from KAIST AI discuss their method ORPO, combining supervised fine-tuning and preference alignment in a single step. They highlight the advantages of their approach, such as minimal data requirement, bias prevention, and enhanced adaptability of language models. The Orpo method has received positive feedback from the research community and industry for efficient alignment and scaling models with smaller datasets.