Preserve-Then-Quantize: Balancing Rank Budgets for Quantization Error Reconstruction in LLMs

Documenti analoghi

• Assigning Distinct Roles to Quantized and Low-Rank Matrices Toward Optimal Weight Decomposition
  di: Cho, Yoonjun, et al.
  Pubblicazione: (2025)
• DAPD: Dependency-Aware Parallel Decoding via Attention for Diffusion LLMs
  di: Kim, Bumjun, et al.
  Pubblicazione: (2026)
• Few-Shot Truly Benign DPO Attack for Jailbreaking LLMs
  di: Yoon, Sangyeon, et al.
  Pubblicazione: (2026)
• Understanding and Mitigating Memorization in Generative Models via Sharpness of Probability Landscapes
  di: Jeon, Dongjae, et al.
  Pubblicazione: (2024)
• Information-Theoretic Discrete Diffusion
  di: Jeon, Moongyu, et al.
  Pubblicazione: (2025)