Yupeng Chang, Yuan Wu 0002, Yi Chang 0001. BV-Blend: Uncertainty-Weighted Historical Baselines for Stable Critic-Free RL with Verifiable Rewards. In Maria Liakata, Viviane P. Moreira, Jiajun Zhang 0001, David Jurgens, editors, Findings of the Association for Computational Linguistics, ACL 2026, San Diego, California, United States, July 2-7, 2026. pages 1706-1725, Association for Computational Linguistics, 2026. [doi]
Abstract is missing.