AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Mimic before Reconstruct: Enhancing Masked Autoencoders with Feature Mimicking

Peng Gao, Ziyi Lin, Renrui Zhang, Rongyao Fang, Hongyang Li, Hongsheng Li, Yu Qiao

IJCV 2023 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning · 27 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 mimic loss on visible encoder tokens (ablation)MAE (200 ep) · ImageNet-1K fine-tuningtop-1 accuracy (%): 83.0 -> 84.7Improved
add focused mimicking (salient tokens) (ablation)MR-MAE · ImageNet-1K fine-tuningtop-1 accuracy (%): 84.7 -> 84.9Improved
add multi-layer fusion (ablation)MR-MAE · ImageNet-1K fine-tuningtop-1 accuracy (%): 84.9 -> 85.0Improved
add masked convolution stages (ablation)MR-MAE · ImageNet-1K fine-tuningtop-1 accuracy (%): 85.0 -> 85.5Improved
add masked convolution stages (ablation)MR-MAE · COCOAP box: 51.6 -> 53.0Improved
apply reconstruction to both tokens instead of mimic before reconstruct (ablation)MR-MAE · ImageNet-1K fine-tuningtop-1 accuracy (%): 85.0 -> 83.3Hurt
add low-level reconstruction to high-level mimicking (ablation)MR-MAE · COCOAP box: 50.9 -> 51.6Improved
use CLIP teacher instead of DINO (ablation)MR-MAE · ImageNet-1K fine-tuningtop-1 accuracy (%): 84.0 -> 85.0Improved
mimic CLIP and DINO jointly instead of CLIP alone (ablation)MR-MAE · ImageNet-1K fine-tuningtop-1 accuracy (%): 85.0 -> 83.8Hurt
ensemble separately trained CLIP and DINO models instead of joint (ablation)MR-MAE · ImageNet-1K fine-tuningtop-1 accuracy (%): 83.8 -> 85.5Improved
pre-train 400 instead of 200 epochs (ablation)MR-MAE · ImageNet-1K fine-tuningtop-1 accuracy (%): 85.5 -> 85.8Improved
pre-train 800 instead of 400 epochs (ablation)MR-MAE · ImageNet-1K fine-tuningtop-1 accuracy (%): 85.8 -> 85.8No clear effect

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