AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Dataset Distillation as Data Compression: A Rate-Utility Perspective

Youneng Bao, Yiping Liu, Zhuo Chen, Yongsheng Liang, Mu Li, Kede Ma

ICCV 2025 · Architectures and efficiency › RNNs and normalization · approach: Dataset distillation and data-efficient training · 1 citations (OpenAlex)

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 changedWhereMeasuredEffect
add joint rate-utility optimization to trajectory matchingTM · ImageNet Nette (128x128, bpc<=192kB)accuracy (%): 51.4 -> 76.5Improved
add joint rate-utility optimization to gradient matchingGM · ImageNet subsets avg, GM loss (bpc<=192kB)accuracy (%): 28.1 -> 49.5Improved
add joint rate-utility optimization to distribution matchingDM · ImageNet subsets avg, DM loss (bpc<=192kB)accuracy (%): 25.7 -> 59.1Improved
TM-RUO versus DDiF averaged over architecturesTM-RUO · ImageNet subsets avg, cross-architecture (bpc<=192kB)accuracy (%): 45.2 -> 50.7Improved
TM-RUO versus DDiF on AlexNet (Nette to Squawk average)TM-RUO · AlexNet cross-architectureaccuracy (%): 49.3 -> 46.7Hurt
TM-RUO versus DDiF on ViT variantTM-RUO · ViT cross-architectureaccuracy (%): 43.5 vs 43.4No clear effect
TM-RUO versus DDiF on Woof (TM loss)TM-RUO · ImageNet Woof (bpc<=192kB)accuracy (%): 42.9 -> 46.0Improved
GM-RUO versus DDiF on WoofGM-RUO · ImageNet Woof, GM lossaccuracy (%): 35.2 -> 33.1Hurt
increase spc from 1 to 102 in TM-RUO (ablation)TM-RUO · ImageNet Netteaccuracy (%): 49.2 -> 76.5Improved
TM-RUO at 94 kB versus best baseline on CIFAR-10TM-RUO · CIFAR-10accuracy (%): 74.0 -> 77.5Improved
TM-RUO at 53 kB versus baseline at 120 kB on CIFAR-100TM-RUO · CIFAR-100accuracy (%): 46.1 -> 49.2Improved
raise storage budget from low to high bpc shifts bit allocation toward explicit latents (ablation)TM-RUO · Nette bit allocationshare of implicit bits: 74% -> 7%Mixed

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy