ExpertWeaver: Unlocking the Inherent MoE in Dense LLMs with GLU Activation Patterns

Documenti analoghi

• Each Rank Could be an Expert: Single-Ranked Mixture of Experts LoRA for Multi-Task Learning
  di: Zhao, Ziyu, et al.
  Pubblicazione: (2025)
• Steering MoE LLMs via Expert (De)Activation
  di: Fayyaz, Mohsen, et al.
  Pubblicazione: (2025)
• ExpertWeave: Efficiently Serving Expert-Specialized Fine-Tuned Adapters at Scale
  di: Shi, Ge, et al.
  Pubblicazione: (2025)
• Grove MoE: Towards Efficient and Superior MoE LLMs with Adjugate Experts
  di: Wu, Haoyuan, et al.
  Pubblicazione: (2025)
• Linear-MoE: Linear Sequence Modeling Meets Mixture-of-Experts
  di: Sun, Weigao, et al.
  Pubblicazione: (2025)