Pith. sign in

REVIEW 2 cited by

On the Diversity and Realism of Distilled Dataset: An Efficient Dataset Distillation Paradigm

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2312.03526 v2 pith:CMC432PV submitted 2023-12-06 cs.CV cs.AIcs.LG

On the Diversity and Realism of Distilled Dataset: An Efficient Dataset Distillation Paradigm

classification cs.CV cs.AIcs.LG
keywords datasetdatasetsdistillationdiversityrealismdatadistilledefficient
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Contemporary machine learning requires training large neural networks on massive datasets and thus faces the challenges of high computational demands. Dataset distillation, as a recent emerging strategy, aims to compress real-world datasets for efficient training. However, this line of research currently struggle with large-scale and high-resolution datasets, hindering its practicality and feasibility. To this end, we re-examine the existing dataset distillation methods and identify three properties required for large-scale real-world applications, namely, realism, diversity, and efficiency. As a remedy, we propose RDED, a novel computationally-efficient yet effective data distillation paradigm, to enable both diversity and realism of the distilled data. Extensive empirical results over various neural architectures and datasets demonstrate the advancement of RDED: we can distill the full ImageNet-1K to a small dataset comprising 10 images per class within 7 minutes, achieving a notable 42% top-1 accuracy with ResNet-18 on a single RTX-4090 GPU (while the SOTA only achieves 21% but requires 6 hours).

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Condensing Large-Scale Datasets Directly with Minimal Information Loss

    cs.CV 2026-07 unverdicted novelty 7.0

    CIM directly aligns data distributions to condense large-scale datasets with minimal information loss, achieving new SOTA results on ImageNet-1K distillation at IPC=10.

  2. Dataset Distillation by Influence Matching

    cs.CV 2026-07 reject novelty 5.0

    Inf-Match distills datasets by matching estimated parameter influence of real and synthetic data, reporting SOTA classification and retrieval, but with an unsupported theoretical core.