Research map / Architectures and efficiency / RNNs and normalization
Sparse Parameterization for Epitomic Dataset Distillation
Xing Wei, Anjia Cao, Funing Yang, Zhiheng Ma
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add feature sparsification with k=96 instead of keeping full features (ablation) | SPEED · CIFAR10 IPC 10, ViT eval | accuracy (%): 50.8 -> 51.8 | Improved |
| reduce k to 24 to meet tighter storage (ablation) | SPEED · CIFAR10 IPC 10, ConvNet eval | accuracy (%): 73.5 -> 68.2 | Hurt |
| reduce k to 12 (ablation) | SPEED · CIFAR10 IPC 10, ConvNet eval | accuracy (%): 73.5 -> 57.8 | Hurt |
| increase FReeNet depth from 1 to 2 blocks (ablation) | SPEED · TinyImageNet IPC 1, ConvNet eval | accuracy (%): 20.7 -> 26.9 | Improved |
| increase FReeNet depth from 2 to 3 blocks (ablation) | SPEED · TinyImageNet IPC 1, ConvNet eval | accuracy (%): 26.9 -> 25.4 | Hurt |
| increase number of SAETs from 32 to 64 (ablation) | SPEED · CIFAR100 IPC 1, ConvNet eval | accuracy (%): 38.5 -> 40.0 | Improved |
| increase number of SAETs from 64 to 96 (ablation) | SPEED · CIFAR100 IPC 1, ConvNet eval | accuracy (%): 40.0 -> 37.3 | Hurt |
| increase SAET dimension from 48 to 96 (ablation) | SPEED · CIFAR100 IPC 1, ConvNet eval | accuracy (%): 38.0 -> 40.0 | Improved |
| increase SAET dimension from 96 to 144 (ablation) | SPEED · CIFAR100 IPC 1, ConvNet eval | accuracy (%): 40.0 -> 35.5 | Hurt |
| increase SAET heads from 3 to 4 (ablation) | SPEED · CIFAR100 IPC 1, ConvNet eval | accuracy (%): 40.0 -> 35.9 | Hurt |
| use 1 SAET head instead of 3 (ablation) | SPEED · CIFAR100 IPC 1, ResNet18 eval | accuracy (%): 29.9 -> 32.2 | Improved |
| plug SPEED into trajectory matching (TM) | TM · CIFAR10 IPC 1, ResNet18/MLP eval | accuracy (%): +23.7 | Improved |
Closest papers by meaning
- Boost Self-Supervised Dataset Distillation via Parameterization, Predefined Augmentation, and Approximation (ICLR 2025 · similarity 0.84)
- Dataset Distillation for Pre-Trained Self-Supervised Vision Models (NeurIPS 2025 · similarity 0.84)
- DELT: A Simple Diversity-driven EarlyLate Training for Dataset Distillation (CVPR 2025 · similarity 0.84)
- Large Scale Dataset Distillation with Domain Shift (ICML 2024 · similarity 0.84)
- Enhancing Dataset Distillation via Non-Critical Region Refinement (CVPR 2025 · similarity 0.83)
- Emphasizing Discriminative Features for Dataset Distillation in Complex Scenarios (CVPR 2025 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
