晚期突破结果：Quamba-SE：用于状态空间模型激活量化的软边量化器
Late Breaking Results: Quamba-SE: Soft-edge Quantizer for Activations in State Space Models
AI总结：
Quamba-SE通过自适应尺度实现状态空间模型激活量化的软边量化，提升模型精度与鲁棒性。
摘要：
我们提出Quamba-SE，一种用于状态空间模型（SSM）激活量化 的软边量化器。与现有方法不同，Quamba-SE采用三种自适应尺度：高精度用于小值，标准尺度用于正常值，低精度用于异常值。这保留了异常值信息而不是硬裁剪，同时保持其他值的精度。我们在Mamba-130M上评估了6个零样本基准。结果表明，Quamba-SE在各个基准上均优于Quamba，单个基准上达到+2.68%的提升，6个数据集的平均准确率提升达+0.83%。
英文摘要：
We propose Quamba-SE, a soft-edge quantizer for State Space Model (SSM) activation quantization. Unlike existing methods, using standard INT8 operation, Quamba-SE employs three adaptive scales: high-precision for small values, standard scale for normal values, and low-precision for outliers. This preserves outlier information instead of hard clipping, while maintaining precision for other values. We evaluate on Mamba- 130M across 6 zero-shot benchmarks. Results show that Quamba- SE consistently outperforms Quamba, achieving up to +2.68% on individual benchmarks and up to +0.83% improvement in the average accuracy of 6 datasets.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
