GM-Skip: Metric-Guided Transformer Block Skipping for Efficient Vision-Language Models

Documenti analoghi

• DeeAD: Dynamic Early Exit of Vision-Language Action for Efficient Autonomous Driving
  di: HU, Haibo, et al.
  Pubblicazione: (2025)
• On-Demand Multi-Task Sparsity for Efficient Large-Model Deployment on Edge Devices
  di: Huang, Lianming, et al.
  Pubblicazione: (2025)
• Skip-Vision: Efficient and Scalable Acceleration of Vision-Language Models via Adaptive Token Skipping
  di: Zeng, Weili, et al.
  Pubblicazione: (2025)
• AD-EE: Early Exiting for Fast and Reliable Vision-Language Models in Autonomous Driving
  di: Huang, Lianming, et al.
  Pubblicazione: (2025)
• Skip-It? Theoretical Conditions for Layer Skipping in Vision-Language Models
  di: Hartman, Max, et al.
  Pubblicazione: (2025)