AI Research · research
PRQuant: Permutation Residual Quantization for Low-Overhead Inference
论文摘要:Accuracy of Low-bit quantization of linear layers is often dominated by a small number of outliers. Although existing methods, such as smoothing, rotation, or residual-based a...
为什么重要
该论文来自一手研究源并通过 3qk7 Research Radar 的新鲜度、来源与研究价值筛选;后续代码、模型权重、复现和相关产品事件可继续关联到该记录。
检测到的变化
- 研究论文进入 3qk7 高价值 Research Radar
一手证据
- arXiv cs.LG · 2026-09-22T22:05:43Z