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

Ähnliche Einträge

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