Shanghaoran Quan

A method for generating quality training data for language model fine-tuning.

Exploring how preference learning improves language model alignment with human expectations.