Papers › Pre-training of Lightweight Vision Transformers on Small Datasets with Minimally Scaled Images

Pre-training of Lightweight Vision Transformers on Small Datasets with Minimally Scaled Images

6 Feb 2024arXiv:2402.03752archive 2025-07-28

Jen Hong Tan

Can a lightweight Vision Transformer (ViT) match or exceed the performance of Convolutional Neural Networks (CNNs) like ResNet on small datasets with small image resolutions? This report demonstrates that a pure ViT can indeed achieve superior performance through pre-training, using a masked auto-encoder technique with minimal image scaling. Our experiments on the CIFAR-10 and CIFAR-100 datasets involved ViT models with fewer than 3.65 million parameters and a multiply-accumulate (MAC) count below 0.27G, qualifying them as 'lightweight' models. Unlike previous approaches, our method attains state-of-the-art performance among similar lightweight transformer-based architectures without significantly scaling up images from CIFAR-10 and CIFAR-100. This achievement underscores the efficiency of our model, not only in handling small datasets but also in effectively processing images close to their original scale.

PaperPDF

Code

No code repository is listed for this paper in the archive or in Syntology's graph.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Image Classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Image Classification CIFAR-10 ViT (lightweight, MAE pretrained) Percentage correct 96.41 #112 of 265 Archive leaderboard report
Image Classification CIFAR-100 ViT (lightweight, MAE pre-trained) PARAMS 3.64M #139 of 211 Archive leaderboard report
Image Classification CIFAR-100 ViT (lightweight, MAE pre-trained) Percentage correct 78.27 #139 of 211 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Absolute Position EncodingsAdamAttentionAverage PoolingBPEConvolutionDense ConnectionsDropoutGlobal Average PoolingKaiming InitializationLabel SmoothingLayer NormalizationLinear LayerMax PoolingMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSoftmaxTransformerVision Transformer

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections