Alignment

Improving LLM Generation with Inverse and Forward Alignment: Reward Modelling, Prompting, Fine‑Tuning, and Inference‑Time Optimisation
Practical techniques for improving LLM generation via inverse & forward alignment.
H. Sun, T. Pouplin, Nicolás Astorga, T. Liu, Mihaela van der Schaar
Dec 10, 2024