$μ$-MoE：测试时剪枝作为微粒度混合专家
$μ$-MoE: Test-Time Pruning as Micro-Grained Mixture-of-Experts
AI总结：
本文提出$μ$-MoE，将测试时激活感知剪枝形式化为微粒度混合专家，通过计算高效的校准实现每个提示的自适应结构化稀疏，以降低推理计算需求。
摘要：
为应对大型基础模型巨大的计算需求，人们引入了无需重训练的激活感知压缩技术。然而，由于这些技术依赖校准数据，对于未知的下游任务可能会出现域偏移。通过计算高效的校准，激活感知剪枝可以针对每个提示自适应地执行，从而在推理时降低复杂度。我们将其形式化为一种微专家混合模型，称为 $μ$-MoE。多项实验表明，$μ$-MoE 能够动态地适应任务/提示相关的结构化稀疏性。
英文摘要：
To tackle the huge computational demand of large foundation models, activation-aware compression techniques without retraining have been introduced. However, since these rely on calibration data, domain shift may arise for unknown downstream tasks. With a computationally efficient calibration, activation-aware pruning can be executed for every prompt adaptively, yet achieving reduced complexity at inference. We formulate it as a mixture of micro-experts, called $μ$-MoE. Several experiments demonstrate that $μ$-MoE can dynamically adapt to task/prompt-dependent structured sparsity on the fly.
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。
