DD-Ranking: Rethinking the Evaluation of Dataset Distillation
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909799237550080 |
|---|---|
| author | Li, Zekai Zhong, Xinhao Khaki, Samir Liang, Zhiyuan Zhou, Yuhao Shi, Mingjia Wang, Ziqiao Zhao, Xuanlei Zhao, Wangbo Qin, Ziheng Wu, Mengxuan Zhou, Pengfei Wang, Haonan Zhang, David Junhao Liu, Jia-Wei Wang, Shaobo Liu, Dai Zhang, Linfeng Li, Guang Wang, Kun Zhu, Zheng Ma, Zhiheng Zhou, Joey Tianyi Lv, Jiancheng Jin, Yaochu Wang, Peihao Zhang, Kaipeng Lyu, Lingjuan Huang, Yiran Akata, Zeynep Deng, Zhiwei Wu, Xindi Cazenavette, George Shang, Yuzhang Cui, Justin Gu, Jindong Zheng, Qian Ye, Hao Wang, Shuo Wang, Xiaobo Yan, Yan Yao, Angela Shou, Mike Zheng Chen, Tianlong Bilen, Hakan Mirzasoleiman, Baharan Kellis, Manolis Plataniotis, Konstantinos N. Wang, Zhangyang Zhao, Bo You, Yang Wang, Kai |
| author_facet | Li, Zekai Zhong, Xinhao Khaki, Samir Liang, Zhiyuan Zhou, Yuhao Shi, Mingjia Wang, Ziqiao Zhao, Xuanlei Zhao, Wangbo Qin, Ziheng Wu, Mengxuan Zhou, Pengfei Wang, Haonan Zhang, David Junhao Liu, Jia-Wei Wang, Shaobo Liu, Dai Zhang, Linfeng Li, Guang Wang, Kun Zhu, Zheng Ma, Zhiheng Zhou, Joey Tianyi Lv, Jiancheng Jin, Yaochu Wang, Peihao Zhang, Kaipeng Lyu, Lingjuan Huang, Yiran Akata, Zeynep Deng, Zhiwei Wu, Xindi Cazenavette, George Shang, Yuzhang Cui, Justin Gu, Jindong Zheng, Qian Ye, Hao Wang, Shuo Wang, Xiaobo Yan, Yan Yao, Angela Shou, Mike Zheng Chen, Tianlong Bilen, Hakan Mirzasoleiman, Baharan Kellis, Manolis Plataniotis, Konstantinos N. Wang, Zhangyang Zhao, Bo You, Yang Wang, Kai |
| contents | In recent years, dataset distillation has provided a reliable solution for data compression, where models trained on the resulting smaller synthetic datasets achieve performance comparable to those trained on the original datasets. To further improve the performance of synthetic datasets, various training pipelines and optimization objectives have been proposed, greatly advancing the field of dataset distillation. Recent decoupled dataset distillation methods introduce soft labels and stronger data augmentation during the post-evaluation phase and scale dataset distillation up to larger datasets (e.g., ImageNet-1K). However, this raises a question: Is accuracy still a reliable metric to fairly evaluate dataset distillation methods? Our empirical findings suggest that the performance improvements of these methods often stem from additional techniques rather than the inherent quality of the images themselves, with even randomly sampled images achieving superior results. Such misaligned evaluation settings severely hinder the development of DD. Therefore, we propose DD-Ranking, a unified evaluation framework, along with new general evaluation metrics to uncover the true performance improvements achieved by different methods. By refocusing on the actual information enhancement of distilled datasets, DD-Ranking provides a more comprehensive and fair evaluation standard for future research advancements. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_13300 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | DD-Ranking: Rethinking the Evaluation of Dataset Distillation Li, Zekai Zhong, Xinhao Khaki, Samir Liang, Zhiyuan Zhou, Yuhao Shi, Mingjia Wang, Ziqiao Zhao, Xuanlei Zhao, Wangbo Qin, Ziheng Wu, Mengxuan Zhou, Pengfei Wang, Haonan Zhang, David Junhao Liu, Jia-Wei Wang, Shaobo Liu, Dai Zhang, Linfeng Li, Guang Wang, Kun Zhu, Zheng Ma, Zhiheng Zhou, Joey Tianyi Lv, Jiancheng Jin, Yaochu Wang, Peihao Zhang, Kaipeng Lyu, Lingjuan Huang, Yiran Akata, Zeynep Deng, Zhiwei Wu, Xindi Cazenavette, George Shang, Yuzhang Cui, Justin Gu, Jindong Zheng, Qian Ye, Hao Wang, Shuo Wang, Xiaobo Yan, Yan Yao, Angela Shou, Mike Zheng Chen, Tianlong Bilen, Hakan Mirzasoleiman, Baharan Kellis, Manolis Plataniotis, Konstantinos N. Wang, Zhangyang Zhao, Bo You, Yang Wang, Kai Computer Vision and Pattern Recognition In recent years, dataset distillation has provided a reliable solution for data compression, where models trained on the resulting smaller synthetic datasets achieve performance comparable to those trained on the original datasets. To further improve the performance of synthetic datasets, various training pipelines and optimization objectives have been proposed, greatly advancing the field of dataset distillation. Recent decoupled dataset distillation methods introduce soft labels and stronger data augmentation during the post-evaluation phase and scale dataset distillation up to larger datasets (e.g., ImageNet-1K). However, this raises a question: Is accuracy still a reliable metric to fairly evaluate dataset distillation methods? Our empirical findings suggest that the performance improvements of these methods often stem from additional techniques rather than the inherent quality of the images themselves, with even randomly sampled images achieving superior results. Such misaligned evaluation settings severely hinder the development of DD. Therefore, we propose DD-Ranking, a unified evaluation framework, along with new general evaluation metrics to uncover the true performance improvements achieved by different methods. By refocusing on the actual information enhancement of distilled datasets, DD-Ranking provides a more comprehensive and fair evaluation standard for future research advancements. |
| title | DD-Ranking: Rethinking the Evaluation of Dataset Distillation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2505.13300 |