Andrew Straiton Hard

Google

Papers on Lattice

Total citations

Topics

h-index

Research focus

Constitutional AI & AI Ethics (1)Interpretability & Mechanistic Interp (1)RLHF & Preference Learning (1)

Frequent co-authors

Wenxuan Zhou (1)Shujian Zhang (1)B. Magdalou (1)John Lambert (1)

Papers (1)

Jul 10, 2025

DeepMindJul 10, 2025·also Google Research

DPO Unchained: Your Training Algorithm is Secretly Disentangled in Human Choice Theory

DPO's success isn't just clever engineering—it's deeply rooted in human choice theory, unlocking a surprisingly flexible framework for preference optimization and justifying many DPO extensions.

Wenxuan Zhou, Shujian Zhang, B. Magdalou +4

Constitutional AI & AI Ethics Interpretability & Mechanistic Interp RLHF & Preference Learning

Search

Andrew Straiton Hard

Research focus

Frequent co-authors

Papers (1)