TTQ: Activation-Aware Test-Time Quantization to Accelerate LLM Inference On The Fly

Documenti analoghi

• Temper and Tilt Lead to SLOP: Reward Hacking Mitigation with Inference-Time Alignment
  di: Wang, Ye, et al.
  Pubblicazione: (2026)
• $μ$-MoE: Test-Time Pruning as Micro-Grained Mixture-of-Experts
  di: Koike-Akino, Toshiaki, et al.
  Pubblicazione: (2025)
• Directional Embedding Smoothing for Robust Vision Language Models
  di: Wang, Ye, et al.
  Pubblicazione: (2026)
• LatentLLM: Attention-Aware Joint Tensor Compression
  di: Koike-Akino, Toshiaki, et al.
  Pubblicazione: (2025)
• AWP: Activation-Aware Weight Pruning and Quantization with Projected Gradient Descent
  di: Liu, Jing, et al.
  Pubblicazione: (2025)