Dynamic Memory Compression: Retrofitting LLMs for Accelerated Inference

Documenti analoghi

• Inference-Time Hyper-Scaling with KV Cache Compression
  di: Łańcucki, Adrian, et al.
  Pubblicazione: (2025)
• The Sparse Frontier: Sparse Attention Trade-offs in Transformer LLMs
  di: Nawrot, Piotr, et al.
  Pubblicazione: (2025)
• KV Cache Transform Coding for Compact Storage in LLM Inference
  di: Staniszewski, Konrad, et al.
  Pubblicazione: (2025)
• Scaling Fine-Grained MoE Beyond 50B Parameters: Empirical Evaluation and Practical Insights
  di: Krajewski, Jakub, et al.
  Pubblicazione: (2025)
• Cross-Lingual and Cross-Cultural Variation in Image Descriptions
  di: Berger, Uri, et al.
  Pubblicazione: (2024)