ad515df27be84538181d1588241fe2a9b4180c8d
Model: lm2445/TABPO_llama3.2_3B_3epoch Source: Original Platform
TAB-PO: Preference Optimization with a Token-Level Adaptive Barrier for Token-Critical Structured Generation
Description
Model: lm2445/TABPO_llama3.2_3B_3epoch Source: Original Platform
TAB-PO: Preference Optimization with a Token-Level Adaptive Barrier for Token-Critical Structured Generation