STaMP: Sequence Transformation and Mixed Precision for Low-Precision Activation Quantization

Documenti analoghi

• HadaNorm: Diffusion Transformer Quantization through Mean-Centered Transformations
  di: Federici, Marco, et al.
  Pubblicazione: (2025)
• Dissecting Quantization Error: A Concentration-Alignment Perspective
  di: Federici, Marco, et al.
  Pubblicazione: (2026)
• Low-Rank Quantization-Aware Training for LLMs
  di: Bondarenko, Yelysei, et al.
  Pubblicazione: (2024)
• FPTQuant: Function-Preserving Transforms for LLM Quantization
  di: van Breugel, Boris, et al.
  Pubblicazione: (2025)
• Mixed-Precision Quantization for Language Models: Techniques and Prospects
  di: Rakka, Mariam, et al.
  Pubblicazione: (2025)