notesum.ai

Published at October 30

COMAL: A Convergent Meta-Algorithm for Aligning LLMs with General Preferences

cs.LG
cs.AI
cs.CL
cs.GT

Released Date: October 30, 2024

Authors: Yixin Liu1, Argyris Oikonomou1, Weiqiang Zheng1, Yang Cai1, Arman Cohan2

Aff.: 1Yale University; 2Allen Institute for AI

Arxiv: http://arxiv.org/abs/2410.23223v1