DMGD: Train-Free Dataset Distillation with Semantic-Distribution Matching in Diffusion Models

2026-05-05 · Qichao Wang, Yunhong Lu, Hengyuan Cao, Junyi Zhang, Min Zhang arxiv

Dataset distillation enables efficient training by distilling the information of large-scale datasets into significantly smaller synthetic datasets. Diffusion based paradigms have emerged in recent years, offering novel perspectives for dataset distillation. However, they typically necessitate additional fine-tuning stages, and effective guidance mechanisms remain underexplored. To address these limitations, we rethink diffusion based dataset distillation and propose a Dual Matching Guided Diffusion (DMGD) framework, centered on efficient training-free guidance. We first establish Semantic Matching via conditional likelihood optimization, eliminating the need for auxiliary classifiers. Furthermore, we propose a dynamic guidance mechanism that enhances the diversity of synthetic data while maintaining semantic alignment. Simultaneously, we introduce an optimal transport (OT) based Distribution Matching approach to further align with the target distribution structure. To ensure efficiency, we develop two enhanced strategies for diffusion based framework: Distribution Approximate Matching and Greedy Progressive Matching. These strategies enable effective distribution matching guidance with minimal computational overhead. Experimental results on ImageNet-Woof, ImageNet-Nette, and ImageNet-1K demonstrate that our training-free approach achieves significant improvements, outperforming state-of-the-art (SOTA) methods requiring additional fine-tuning by average accuracy gains of 2.1%, 5.4%, and 2.4%, respectively.

📄 PDF Abstract BibTeX arXiv:2605.03877

Code (0)

등록된 구현이 없습니다.

Federated Learning Meets Multi-objective Optimization

2020-06-20 · Zeou Hu, Kiarash Shaloudegi, Guojun Zhang, Yao-Liang Yu

Federated learning has emerged as a promising, massively distributed way to train a joint deep model over large amounts of edge devices while keeping private user data strictly on device. In this work, motivated from ens…

Fairness Federated Learning

Modeling Global Distribution for Federated Learning with Label Distribution Skew

2022-12-17 · Tao Sheng, Chengchao Shen, YuAn Liu, Yeyu Ou 외

Federated learning achieves joint training of deep models by connecting decentralized data sources, which can significantly mitigate the risk of privacy leakage. However, in a more general case, the distributions of labe…

Federated Learning Generative Adversarial Network

Integrating Network Embedding and Community Outlier Detection via Multiclass Graph Description

2020-07-20 · Sambaran Bandyopadhyay, Saley Vishal Vivek, M. N. Murty

Network (or graph) embedding is the task to map the nodes of a graph to a lower dimensional vector space, such that it preserves the graph properties and facilitates the downstream network mining tasks. Real world networ…

Community Detection Graph Embedding Network Embedding Outlier Detection

ManifoldGD: Training-Free Hierarchical Manifold Guidance for Diffusion-Based Dataset Distillation

2026-02-26 · Ayush Roy, Wei-Yang Alex Lee, Rudrasis Chakraborty, Vishnu Suresh Lokhande arxiv

In recent times, large datasets hinder efficient model training while also containing redundant concepts. Dataset distillation aims to synthesize compact datasets that preserve the knowledge of large-scale training sets …

Seeing the Whole Picture: Distribution-Guided Data-Free Distillation for Semantic Segmentation

2025-12-15 · Hongxuan Sun, Tao Wu arxiv

Semantic segmentation requires a holistic understanding of the physical world, as it assigns semantic labels to spatially continuous and structurally coherent objects rather than to isolated pixels. However, existing dat…

Data-free Knowledge Distillation Semantic Segmentation