← 返回专题广场
direct-preference-optimization
1 个项目 · ⭐ 151
[ICLR 2026] Official repository of "Uni-DPO: A Unified Paradigm for Dynamic Preference Optimization of LLMs".
Python
⭐ 15
⑂ 0
Apache-2.0
· 2026-05-16推送
2026-05-16
最近推送