ArmoRM
收藏官方服务:
资源简介:
Absolute-Rating Multi-Objective Reward Model (ArmoRM) with Mixture-of-Experts (M
绝对评价多目标奖励模型(Absolute-Rating Multi-Objective Reward Model,简称ArmoRM)与专家混合模型(Mixture-of-Experts,简称MoE)
创建时间:
2024-10-17

Absolute-Rating Multi-Objective Reward Model (ArmoRM) with Mixture-of-Experts (M
绝对评价多目标奖励模型(Absolute-Rating Multi-Objective Reward Model,简称ArmoRM)与专家混合模型(Mixture-of-Experts,简称MoE)