HESTIA: A Hessian-Guided Differentiable Quantization-Aware Training Framework for Extremely Low-Bit LLMs

Ähnliche Einträge

• Fairy2i: Training Complex LLMs from Real LLMs with All Parameters in $\{\pm 1, \pm i\}$
  von: Wang, Feiyu, et al.
  Veröffentlicht: (2025)
• What Makes Low-Bit Quantization-Aware Training Work for Reasoning LLMs? A Systematic Study
  von: Lv, Keyu, et al.
  Veröffentlicht: (2026)
• HeRo-Q: A General Framework for Stable Low Bit Quantization via Hessian Conditioning
  von: Zhang, Jinhao Zhang Yunquan, et al.
  Veröffentlicht: (2026)
• PTQ1.61: Push the Real Limit of Extremely Low-Bit Post-Training Quantization Methods for Large Language Models
  von: Zhao, Jiaqi, et al.
  Veröffentlicht: (2025)
• Quant-dLLM: Post-Training Extreme Low-Bit Quantization for Diffusion Large Language Models
  von: Zhang, Tianao, et al.
  Veröffentlicht: (2025)