AI Paper Atlas

Research map / Label-efficient and robust learning / Knowledge distillation

Sequential Subset Matching for Dataset Distillation

Jiawei Du, Qin Shi, Joey Tianyi Zhou

NeurIPS 2023 · Label-efficient and robust learning › Knowledge distillation · approach: Self and data-free distillation · 5 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
plug SeqMatch into MTTMTT · CIFAR-100 ipc=50test accuracy (%): 47.7 -> 51.2Improved
plug SeqMatch into MTTMTT · CIFAR-10 ipc=50test accuracy (%): 71.6 -> 74.4Improved
plug SeqMatch into MTTMTT · CIFAR-10 ipc=10test accuracy (%): 65.3 -> 66.2Improved
plug SeqMatch into MTTMTT · Tiny ImageNet ipc=10test accuracy (%): 23.2 -> 23.8Improved
plug SeqMatch into IDCIDC · SVHN ipc=50test accuracy (%): 90.2 -> 92.1Improved
plug SeqMatch into IDCIDC · SVHN ipc=10test accuracy (%): 87.3 -> 87.2No clear effect
plug SeqMatch into IDCIDC · CIFAR-10 ipc=50test accuracy (%): 74.5 -> 75.3Improved
plug SeqMatch into MTTMTT · MNIST ipc=50test accuracy (%): 98.5 -> 99.0Improved
evaluate SeqMatch-MTT on other architecturesMTT · CIFAR-10 ipc=50, ResNet18test accuracy (%): 61.9 -> 68.4Improved
evaluate SeqMatch-MTT on other architecturesMTT · CIFAR-10 ipc=50, VGG11test accuracy (%): 55.4 -> 64.2Improved
evaluate SeqMatch-MTT on other architecturesMTT · CIFAR-10 ipc=50, AlexNettest accuracy (%): 48.2 -> 50.7Improved
plug SeqMatch into MTTMTT · ImageNette ipc=10test accuracy (%): 63.0 -> 66.9Improved

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