Skip to content
TL;DR
Derive a closed-form loss that optimizes a policy against preference data without training a separate reward model or running PPO. Much simpler than RLHF, competitive quality.
BACKLOG · WORK IN PROGRESS

This paper is being written.

The metadata and shape of this page are stable, but the body content isn't ready yet. We'll publish it once it meets the bar of teaching something new with worked examples and real tools.