PagedEviction: Structured Block-wise KV Cache Pruning for Efficient Large Language Model Inference

Documenti analoghi

• LangVision-LoRA-NAS: Neural Architecture Search for Variable LoRA Rank in Vision Language Models
  di: Chitty-Venkata, Krishna Teja, et al.
  Pubblicazione: (2025)
• LExI: Layer-Adaptive Active Experts for Efficient MoE Model Inference
  di: Chitty-Venkata, Krishna Teja, et al.
  Pubblicazione: (2025)
• BaKlaVa -- Budgeted Allocation of KV cache for Long-context Inference
  di: Gulhan, Ahmed Burak, et al.
  Pubblicazione: (2025)
• ImageNet-Think-250K: A Large-Scale Synthetic Dataset for Multimodal Reasoning for Vision Language Models
  di: Chitty-Venkata, Krishna Teja, et al.
  Pubblicazione: (2025)
• MoPEQ: Mixture of Mixed Precision Quantized Experts
  di: Chitty-Venkata, Krishna Teja, et al.
  Pubblicazione: (2025)