通过动态选图与拓扑对齐，提升数据蒸馏的类别多样性与泛化能力。
Fixed Anchors Are Not Enough: Dynamic Retrieval and Persistent Homology for Dataset Distillation
- 动态检索真实图像块注入合成过程，平衡拟合与复杂度。
- 用持久同调分析真实与合成数据的拓扑结构差异，减少类内相似性缺陷。
- 在多个数据集上超越已有方法，尤其在ImageNet-1K上达64.3%准确率。
解耦式数据蒸馏（DD）通过匹配冻结教师模型的统计特征，将大规模数据集压缩为少量合成图像。然而，现有残差匹配流程依赖静态真实图像块，导致拟合-复杂度差距和拉向锚点效应，降低类内多样性并损害泛化性能。为此，我们提出RET A——一种用于解耦式数据蒸馏的检索与拓扑对齐框架。首先，动态检索连接（DRC）从预构建池中选择真实图像块，通过最小化教师特征空间中的拟合-复杂度评分实现注入，以增强特征拟合同时控制引入复杂度。其次，持久同调对齐（PTA）通过构建互近邻特征图，计算组件与环路的持久性图像，惩罚真实与合成数据集间的拓扑差异，缓解拉向锚点效应。在CIFAR-100、Tiny-ImageNet、ImageNet-1K及多个ImageNet子集上，RETA始终优于多种基线，在使用ResNet-18时，于每类50张图像条件下，使ImageNet-1K的Top-1准确率达到64.3%，较最优前人方法提升3.1%。
原文摘要 · Abstract (English)
Decoupled dataset distillation (DD) compresses large corpora into a few synthetic images by matching a frozen teacher's statistics. However, current residual-matching pipelines rely on static real patches, creating a fit-complexity gap and a pull-to-anchor effect that reduce intra-class diversity and hurt generalization. To address these issues, we introduce RETA -- a Retrieval and Topology Alignment framework for decoupled DD. First, Dynamic Retrieval Connection (DRC) selects a real patch from a prebuilt pool by minimizing a fit-complexity score in teacher feature space; the chosen patch is injected via a residual connection to tighten feature fit while controlling injected complexity. Second, Persistent Topology Alignment (PTA) regularizes synthesis with persistent homology: we build a mutual k-NN feature graph, compute persistence images of components and loops, and penalize topology discrepancies between real and synthetic sets, mitigating pull-to-anchor effect. Across CIFAR-100, Tiny-ImageNet, ImageNet-1K, and multiple ImageNet subsets, RETA consistently outperforms various baselines under comparable time and memory, especially reaching 64.3% top-1 accuracy on ImageNet-1K with ResNet-18 at 50 images per class, +3.1% over the best prior.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。