Yuexiang Xie

Latest

• $\beta$-DPO: Direct Preference Optimization with Dynamic $\beta$