Cross-Self KV Cache Pruning for Efficient Vision-Language Inference

Documenti analoghi

• Rethinking Causal Mask Attention for Vision-Language Inference
  di: Pei, Xiaohuan, et al.
  Pubblicazione: (2025)
• EfficientVMamba: Atrous Selective Scan for Light Weight Visual Mamba
  di: Pei, Xiaohuan, et al.
  Pubblicazione: (2024)
• AirCache: Activating Inter-modal Relevancy KV Cache Compression for Efficient Large Vision-Language Model Inference
  di: Huang, Kai, et al.
  Pubblicazione: (2025)
• KV-Efficient VLA: A Method to Speed up Vision Language Models with RNN-Gated Chunked KV Cache
  di: Xu, Wanshun, et al.
  Pubblicazione: (2025)
• KVCapsule: Efficient Sequential KV Cache Compression for Vision-Language Models with Asymmetric Redundancy
  di: Huang, Yingbing, et al.
  Pubblicazione: (2026)