OCNR: Stabilizing Self-Play by Mitigating Iteration-Collapse With One-Class Novelty Rewards

Jul 6, 2026·
Seungyoo Lee
Seungyoo Lee
,
Giung Nam
,
Hyungi Lee†
,
Juho Lee†
· 0 min read
Type
Publication
In Proceedings of the 43rd International Conference on Machine Learning (ICML 2026), PMLR 306
Accepted
April 30, 2026