Sharp Analysis for KL-Regularized Contextual Bandits and RLHF

Open in new window