eXmY: A Data Type and Technique for Arbitrary Bit Precision Quantization

Documenti analoghi

• Accurate Block Quantization in LLMs with Outliers
  di: Trukhanov, Nikita, et al.
  Pubblicazione: (2024)
• Bit-Accurate Modeling of GPU Matrix Multiply-Accumulate Units: Demystifying Numerical Discrepancy and Accuracy
  di: Xie, Peichen, et al.
  Pubblicazione: (2025)
• An Open-Source Framework for Efficient Numerically-Tailored Computations
  di: Ledoux, Louis, et al.
  Pubblicazione: (2024)
• APT-LLM: Exploiting Arbitrary-Precision Tensor Core Computing for LLM Acceleration
  di: Ma, Shaobo, et al.
  Pubblicazione: (2025)
• Efficient Arbitrary Precision Acceleration for Large Language Models on GPU Tensor Cores
  di: Ma, Shaobo, et al.
  Pubblicazione: (2024)