FastMMoE: Accelerating Multimodal Large Language Models through Dynamic Expert Activation and Routing-Aware Token Pruning

Documenti analoghi

• SMAR: Soft Modality-Aware Routing Strategy for MoE-based Multimodal Large Language Models Preserving Language Capabilities
  di: Xia, Guoyang, et al.
  Pubblicazione: (2025)
• MMoE: Enhancing Multimodal Models with Mixtures of Multimodal Interaction Experts
  di: Yu, Haofei, et al.
  Pubblicazione: (2023)
• ChunkLLM: A Lightweight Pluggable Framework for Accelerating LLMs Inference
  di: Ouyang, Haojie, et al.
  Pubblicazione: (2025)
• FastVID: Dynamic Density Pruning for Fast Video Large Language Models
  di: Shen, Leqi, et al.
  Pubblicazione: (2025)
• QAPruner: Quantization-Aware Vision Token Pruning for Multimodal Large Language Models
  di: Wang, Xinhao, et al.
  Pubblicazione: (2026)