TAB-PO: Preference Optimization with a Token-Level Adaptive Barrier for Token-Critical Structured Generation

https://arxiv.org/abs/2603.00025

https://github.com/SarielMa/TAB-PO