MXFormer: A Microscaling Floating-Point Charge-Trap Transistor Compute-in-Memory Transformer Accelerator

Documenti analoghi

• CIMPool: Scalable Neural Network Acceleration for Compute-In-Memory using Weight Pools
  di: Li, Shurui, et al.
  Pubblicazione: (2025)
• MXDOTP: A RISC-V ISA Extension for Enabling Microscaling (MX) Floating-Point Dot Products
  di: İslamoğlu, Gamze, et al.
  Pubblicazione: (2025)
• TimeFloats: Train-in-Memory with Time-Domain Floating-Point Scalar Products
  di: Hashem, Maeesha Binte, et al.
  Pubblicazione: (2024)
• CATCH: a Cost Analysis Tool for Co-optimization of chiplet-based Heterogeneous systems
  di: Graening, Alexander, et al.
  Pubblicazione: (2025)
• A Hybrid-Domain Floating-Point Compute-in-Memory Architecture for Efficient Acceleration of High-Precision Deep Neural Networks
  di: Yi, Zhiqiang, et al.
  Pubblicazione: (2025)