MoE-LLaVA: Mixture of Experts for Large Vision-Language Models

Documenti analoghi

• Video-LLaVA: Learning United Visual Representation by Alignment Before Projection
  di: Lin, Bin, et al.
  Pubblicazione: (2023)
• LLaVA-MoD: Making LLaVA Tiny via MoE Knowledge Distillation
  di: Shu, Fangxun, et al.
  Pubblicazione: (2024)
• LLaVA-CMoE: Towards Continual Mixture of Experts for Large Vision-Language Models
  di: Zhao, Hengyuan, et al.
  Pubblicazione: (2025)
• ATP-LLaVA: Adaptive Token Pruning for Large Vision Language Models
  di: Ye, Xubing, et al.
  Pubblicazione: (2024)
• LLaVA-CoT: Let Vision Language Models Reason Step-by-Step
  di: Xu, Guowei, et al.
  Pubblicazione: (2024)