IMS3：打破基于扩散的数据集蒸馏中的分布聚合
IMS3: Breaking Distributional Aggregation in Diffusion-Based Dataset Distillation
该论文提出反演引导的微调与选择性采样，可提升扩散模型生成样本的判别性与多样性，对需要合成训练数据的视觉任务（如分类、检测）有直接参考价值。
原文依据 diffusion models are optimized for generative likelihood rather than discriminative utility
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论数据集蒸馏旨在合成紧凑数据集，以近似大规模真实数据集的训练效果，为现代深度学习日益增长的计算需求提供高效解决方案。近年来，基于扩散的数据集蒸馏方法通过利用扩散模型强大的生成能力产生多样且结构一致的样本，展现出巨大潜力。然而，一个根本性的目标错位依然存在：扩散模型优化的是生成似然而非判别效用，导致样本过度集中于高密度区域，对分类至关重要的边界样本覆盖不足。为解决该问题，我们提出两种互补策略。反演匹配（IM）引入反演引导的微调过程，使去噪轨迹与其反演对应轨迹对齐，拓宽分布覆盖并增强多样性。选择性子群采样（S^3）是一种无需训练的采样机制，通过选择兼具代表性与区分性的合成子集来提升类间可分性。大量实验表明，我们的方法显著提升了蒸馏数据集的判别质量与泛化能力，在基于扩散的方法中达到最优性能。
英文原文摘要 1370 字符
Dataset Distillation aims to synthesize compact datasets that can approximate the training efficacy of large-scale real datasets, offering an efficient solution to the increasing computational demands of modern deep learning. Recently, diffusion-based dataset distillation methods have shown great promise by leveraging the strong generative capacity of diffusion models to produce diverse and structurally consistent samples. However, a fundamental goal misalignment persists: diffusion models are optimized for generative likelihood rather than discriminative utility, resulting in over-concentration in high-density regions and inadequate coverage of boundary samples crucial for classification. To address this issue, we propose two complementary strategies. Inversion-Matching (IM) introduces an inversion-guided fine-tuning process that aligns denoising trajectories with their inversion counterparts, broadening distributional coverage and enhancing diversity. Selective Subgroup Sampling( S^3 ) is a training-free sampling mechanism that improves inter-class separability by selecting synthetic subsets that are both representative and distinctive. Extensive experiments demonstrate that our approach significantly enhances the discriminative quality and generalization of distilled datasets, achieving state-of-the-art performance among diffusion-based methods.