DistrAttention: An Efficient and Flexible Self-Attention Mechanism on Modern GPUs

Documenti analoghi

• SpecPipe: Accelerating Pipeline Parallelism-based LLM Inference with Speculative Decoding
  di: Yin, Haofei, et al.
  Pubblicazione: (2025)
• TAMO: Fine-Grained Root Cause Analysis via Tool-Assisted LLM Agent with Multi-Modality Observation Data in Cloud-Native Systems
  di: Zhang, Xiao, et al.
  Pubblicazione: (2025)
• COMCAT: Towards Efficient Compression and Customization of Attention-Based Vision Models
  di: Xiao, Jinqi, et al.
  Pubblicazione: (2023)
• Revealing the Attention Floating Mechanism in Masked Diffusion Models
  di: Dai, Xin, et al.
  Pubblicazione: (2026)
• Data-Free Continual Learning of Server Models in Model-Heterogeneous Cloud-Device Collaboration
  di: Zhang, Xiao, et al.
  Pubblicazione: (2025)