Seonglae ChoOut-of-Distribution
When data distributions differ, Transfer Learning or Meta Learning helps adapt models to new tasks.
OOD generalisation is crucial given the wide range of real-world scenarios in which these models are being used, while output diversity refers to the model’s ability to generate varied outputs and is important for a variety of use cases
RLHF generalises better than SFT to new inputs, particularly as the distribution shift between train and test becomes larger. However, RLHF significantly reduces output diversity compared to SFT across a variety of measures, implying a tradeoff in current LLM fine-tuning methods between generalisation and diversity.