AI 中文总结
本文提出统一响应几何用于结构化剪枝，通过联合响应容量选择子集并补偿移除响应，在ImageNet上显著提升剪枝后准确率，验证了响应几何作为条件原则的有效性。
AI 中文摘要
结构化剪枝通常被表述为对单个通道进行排序，尽管通道响应可能通过下游混合而互补或抵消。受这些响应相互作用的启发，我们将剪枝表述为选择具有较大联合响应容量的子集，随后进行单独的功能实现步骤。我们的统一响应几何将每个候选集映射为\(M(D,R)=D^{1/2}RD^{1/2}\)，并利用其行列式以及Schur-贪婪残差来选择非冗余坐标。相同的构造产生了两个信息条件实例：一个基于激活协方差的无标签实例，以及一个任务条件实例，该实例结合激活和梯度方差用于响应尺度，并结合梯度相关性用于互补性。为了将选定的子集转换为可执行的网络，我们通过岭补偿将可预测的移除响应折叠到后继权重中，并重新校准批归一化统计量，而无需对网络进行微调。在ImageNet ResNet-50上，无标签实例在30%和40%删除率下分别达到65.4%和53.9%的Top-1准确率，而仅基于强度的选择分别为59.8%和43.1%；任务条件实例在同一协议下分别达到67.7%和56.3%。一个六族筛选显示了依赖于架构的行为，在几种卷积和扩展层设置中具有正相对对比度，在窗口注意力中具有明确的边界情况。这些结果支持响应几何作为结构化剪枝的条件原则，其收益由观察到的响应和实现该响应的架构共同决定。
英文摘要
Structured pruning is commonly formulated as ranking individual channels, although channel responses can be complementary or cancel through downstream mixing. Motivated by these response interactions, we formulate pruning as the selection of a subset with large joint response capacity, followed by a separate functional realization step. Our unified response geometry maps each candidate set to \(M(D,R)=D^{1/2}RD^{1/2}\) and uses its determinant together with Schur-greedy residuals to select non-redundant coordinates. The same construction yields two information-conditioned instances: an unlabeled instance based on activation covariance, and a task-conditioned instance that combines activation and gradient variance for response scale with gradient correlation for complementarity. To convert the selected subset into an executable network, we fold predictable removed responses into successor weights through ridge compensation and recalibrate batch-normalization statistics, without fine-tuning the network. On ImageNet ResNet-50, the unlabeled instance reaches \(65.4\%\) and \(53.9\%\) Top-1 accuracy at 30\% and 40\% deletion, versus \(59.8\%\) and \(43.1\%\) for strength-only selection; the task-conditioned instance reaches \(67.7\%\) and \(56.3\%\) under the same protocol. A six-family screen shows architecture-dependent behavior, with positive relative contrasts in several convolutional and expansion-layer settings and clear boundary cases in windowed attention. These results support response geometry as a conditional principle for structured pruning, with its benefit determined jointly by the observed response and the architecture in which that response is realized.
