WANG, Chenglong et al. Probing Preference Representations: A Multi-Dimensional Evaluation and Analysis Method for Reward Models. Proceedings of the AAAI Conference on Artificial Intelligence, [S. l.], v. 40, n. 39, p. 33404–33412, 2026. DOI: 10.1609/aaai.v40i39.40627. Disponível em: https://ojs.aaai.org/index.php/AAAI/article/view/40627. Acesso em: 16 may. 2026.