Mosaic: Unlocking Long-Context Inference for Diffusion LLMs via Global Memory Planning and Dynamic Peak Taming

Documenti analoghi

• Harpagon: Minimizing DNN Serving Cost via Efficient Dispatching, Scheduling and Splitting
  di: Zhao, Zhixin, et al.
  Pubblicazione: (2024)
• PAT: Accelerating LLM Decoding via Prefix-Aware Attention with Resource Efficient Multi-Tile Kernel
  di: Yi, Jinjun, et al.
  Pubblicazione: (2025)
• LongLLaDA: Unlocking Long Context Capabilities in Diffusion LLMs
  di: Liu, Xiaoran, et al.
  Pubblicazione: (2025)
• Taming Wild Knots with Mosaics
  di: Deng, Mary Y., et al.
  Pubblicazione: (2026)
• RAGPulse: An Open-Source RAG Workload Trace to Optimize RAG Serving Systems
  di: Wang, Zhengchao, et al.
  Pubblicazione: (2025)