Research map / Label-efficient and robust learning / Recognition and long-tail
TGDD: Trajectory Guided Dataset Distillation with Balanced Distribution
Fengli Ran, Xiao Pu, Bo Liu, Xiuli Bi, Bin Xiao
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 multiformation augmentation (ablation) | TGDD · CIFAR-10 IPC-10 | accuracy: 55.2 -> 63.2 | Improved |
| add stage-wise distribution matching over expert trajectories (ablation) | TGDD · CIFAR-10 IPC-10 | accuracy: 63.2 -> 65.8 | Improved |
| add stage-wise distribution constraint (ablation) | TGDD · CIFAR-10 IPC-10 | accuracy: 65.8 -> 71.9 | Improved |
| add stage-wise distribution constraint (ablation) | TGDD · CIFAR-10 IPC-50 | accuracy: 75.2 -> 76.5 | Improved |
| add stage-wise distribution constraint (ablation) | TGDD · CIFAR-100 IPC-10 | accuracy: 47.0 -> 51.3 | Improved |
| vary regularization coefficient alpha 0.5 to 10.5 (ablation) | TGDD · IPC-10 | accuracy: fluctuates within 2.8% | No clear effect |
| use expert region of snapshots instead of single converged expert (ablation) | TGDD · IPC-10 | accuracy | Improved |
| increase number or length of expert trajectories (ablation) | TGDD · IPC-10 | accuracy | No clear effect |
| TGDD vs DM on CIFAR-10 | TGDD · CIFAR-10 | accuracy: 48.9 -> 71.9 (IPC-10) | Improved |
| use TGDD synthetic sets on ResNet-10 | TGDD · CIFAR-10 IPC-50 | accuracy: DANCE 68.0 -> 74.9 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- DSDM — DsDm: Model-Aware Dataset Selection with Datamodels (ICML 2024) · better on 4 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Towards Lossless Dataset Distillation via Difficulty-Aligned Trajectory Matching (ICLR 2024 · similarity 0.90)
- Towards Stable and Storage-efficient Dataset Distillation: Matching Convexified Trajectory (CVPR 2025 · similarity 0.90)
- Path-Guided Flow Matching for Dataset Distillation (NeurIPS 2026 · similarity 0.87)
- Optimizing Distributional Geometry Alignment with Optimal Transport for Generative Dataset Distillation (NeurIPS 2025 · similarity 0.87)
- Distilling Long-tailed Datasets (CVPR 2025 · similarity 0.86)
- Dataset Distillation of 3D Point Clouds via Distribution Matching (NeurIPS 2025 · similarity 0.86)
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.
