tracked author

Fuli Luo (罗福莉)

MOPD co-corresponding author and Xiaomi MiMo lead signal. Google Scholar identifies Fuli Luo(罗福莉); OpenReview and Hugging Face activity connect her to MiMo-V2-Flash, MoE RL router alignment, DeepSeek-R1 and prior DeepSeek work. X search result and posts from @_LuoFuli discuss MiMo releases, matching Xiaomi MiMo author-line evidence. PKU news article describes Luo Fuli as leader of the MiMo LLM. X account confidence is high because the handle, MiMo posts, public profile snippets, paper activity and external reporting align.

2 archived notes X: high HomepageX

Related Notes

按论文归档时间排序,展示该作者在本站已经出现的材料。

Jul 02, 2026

MOPD: Multi Teacher On Policy Distillation for Capability Integration in LLM Post Training

MOPD 把“多个领域 RL teacher 的能力整合”改写成一个 on policy token level distillation 问题:student 先用自己的当前策略生成轨迹,再让对应领域 teacher 在同一轨迹前缀上做 teacher forced prefill,最后用 reverse KL 或等价 policy gradient advantage 把 teacher 的局部偏好注入 student;关键成立条...

2606.30406-mopd-multi-teacher-on-policy-distillation On-Policy DistillationRL AlgorithmReasoning RL
Jun 24, 2026

DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models

DeepSeekMoE 把 MoE (Mixture of Experts,混合专家) 的效率问题改写成专家专门化问题:fine grained expert segmentation 通过切小 FFN (Feed Forward Network,前馈网络) 专家并增加激活专家数,提高每个 token 的专家组合分辨率;shared expert isolation 通过固定激活共享专家承载通用知识,让 routed experts ...

2401.06066-deepseekmoe-expert-specialization MoE Architecture