Shangmin Guo

DeRa offers a method to adjust language model alignment without retraining.

A new method improves AI alignment using real-time feedback.

This paper discusses how language models learn and evolve through interaction.