Grove MoE: Towards Efficient and Superior MoE LLMs with Adjugate Experts
Paper
•
2508.07785
•
Published
•
25
MoE
-
-
MoBE: Mixture-of-Basis-Experts for Compressing MoE-based LLMs
Paper • 2508.05257 • Published • 12 -
SmallThinker: A Family of Efficient Large Language Models Natively Trained for Local Deployment
Paper • 2507.20984 • Published • 56 -
MiniCPM4: Ultra-Efficient LLMs on End Devices
Paper • 2506.07900 • Published • 90