Large Language Model

This blog post notes my understanding of the algorithm DPO (Direct Preference Optimization) and the math derivation behind it.