DuQuant++: Fine-grained Rotation Enhances Microscaling FP4 Quantization

Documenti analoghi

• DuQuant: Distributing Outliers via Dual Transformation Makes Stronger Quantized LLMs
  di: Lin, Haokun, et al.
  Pubblicazione: (2024)
• Quantization Meets dLLMs: A Systematic Study of Post-training Quantization for Diffusion LLMs
  di: Lin, Haokun, et al.
  Pubblicazione: (2025)
• TokLIP: Marry Visual Tokens to CLIP for Multimodal Comprehension and Generation
  di: Lin, Haokun, et al.
  Pubblicazione: (2025)
• LRQ-DiT: Log-Rotation Post-Training Quantization of Diffusion Transformers for Image and Video Generation
  di: Yang, Lianwei, et al.
  Pubblicazione: (2025)
• DopQ-ViT: Towards Distribution-Friendly and Outlier-Aware Post-Training Quantization for Vision Transformers
  di: Yang, Lianwei, et al.
  Pubblicazione: (2024)