Reward-free Alignment for Conflicted Objectives

RACO has been accepted to ICML 2026 as Oral paper (Top 0.7%)

This repo contains set of models with safe usage for paper -

Reward-free Alignment for Conflicting Objectives

https://arxiv.org/abs/2602.02495

Please cite

@inproceedings{
chen2026rewardfree,
title={Reward-free Alignment for Conflicting Objectives},
author={Peter Chen and Xiaopeng Li and Xi Chen and Tianyi Lin}
booktitle={Forty-third International Conference on Machine Learning},
year={2026},
url={https://openreview.net/forum?id=vSzRJyg6k0}
}