Q$^2$: Quantization-Aware Gradient Balancing and Attention Alignment for Low-Bit Quantization

Documenti analoghi

• MimiQ: Low-Bit Data-Free Quantization of Vision Transformers with Encouraging Inter-Head Attention Similarity
  di: Choi, Kanghyun, et al.
  Pubblicazione: (2024)
• Reclaiming Residual Knowledge: A Novel Paradigm to Low-Bit Quantization
  di: Luo, Róisín, et al.
  Pubblicazione: (2024)
• Fine-Grained Post-Training Quantization for Large Vision Language Models with Quantization-Aware Integrated Gradients
  di: Xiang, Ziwei, et al.
  Pubblicazione: (2026)
• Breaking Modality Heterogeneity in Low-Bit Quantization for Large Vision-Language Models
  di: Zhong, Yi, et al.
  Pubblicazione: (2026)
• Quantization Variation: A New Perspective on Training Transformers with Low-Bit Precision
  di: Huang, Xijie, et al.
  Pubblicazione: (2023)