AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

Simple Filtering Improves Masked Autoencoders

Takumi Kobayashi

ECCV 2026 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Masked autoencoder pretraining · 0 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
replace random masking with 2x2 max-pool filtered masking (ablation)MAE ViT-S · ImageNet-100accuracy (%): 68.99 -> 72.07Improved
replace random masking with 2x2 max-pool filtered masking (ablation)MAE ViT · Cifar-10accuracy (%): 79.18 -> 82.16Improved
use 4x4 max-pool kernel instead of 2x2 (ablation)MAE ViT-S · ImageNet-100accuracy (%): 72.07 -> 69.93Hurt
replace random masking with grid masking (ablation)MAE ViT-S · ImageNet-100accuracy (%): 68.99 -> 58.82Hurt
replace random masking with Laplacian filtered masking (ablation)MAE ViT-S · ImageNet-100accuracy (%): 68.99 -> 66.68Hurt
smooth reconstruction target with Gaussian sigma=4 (ablation)MAE ViT-S random masking · ImageNet-100accuracy (%): 68.99 -> 72.10Improved
smooth reconstruction target with Gaussian sigma=4 on top of max-pool masking (ablation)MAE ViT-S · ImageNet-100accuracy (%): 72.07 -> 73.29Improved
sharpen target (unsharp masking) instead of smoothing (ablation)MAE ViT-S · ImageNet-100accuracy (%): 72.07 -> 64.51Hurt
Gaussian smoothing instead of down-scaling target (ablation)MAE ViT-S max-pool · ImageNet-100accuracy (%): 72.45 -> 73.29Improved
smoothing with sigma=8 vs max-pool alone (ablation)MAE ViT-S max-pool · ImageNet-100accuracy (%): 72.07 -> 72.07No clear effect
use filtering method instead of frequency losses (ablation)MAE ViT-S · ImageNet-100accuracy (%): 70.94 -> 73.29Improved
apply method to MAE ViT-B at 1600 epochsMAE ViT-B · ImageNet-1K linear probeaccuracy (%): 66.71 -> 72.55Improved

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