Anda: Unlocking Efficient LLM Inference with a Variable-Length Grouped Activation Data Format

Documenti analoghi

• Fine-Grained Fusion: The Missing Piece in Area-Efficient State Space Model Acceleration
  di: Geens, Robin, et al.
  Pubblicazione: (2025)
• Hardware-Centric Analysis of DeepSeek's Multi-Head Latent Attention
  di: Geens, Robin, et al.
  Pubblicazione: (2025)
• Hardware Generation and Exploration of Lookup Table-Based Accelerators for 1.58-bit LLM Inference
  di: Geens, Robin, et al.
  Pubblicazione: (2026)
• The Hyperscale Lottery: How State-Space Models Have Sacrificed Edge Efficiency
  di: Geens, Robin, et al.
  Pubblicazione: (2026)
• Precision-Scalable Microscaling Datapaths with Optimized Reduction Tree for Efficient NPU Integration
  di: Cuyckens, Stef, et al.
  Pubblicazione: (2025)