Rejection sampling fine-tuning

Pairings in the atlas

Best-of-n distillationspecialistPreference alignment llm-training-alignment

Rivals: other methods for the same problems

Where it sits

llm-training-alignment · Machine Learning & AI