# Mimic before Reconstruct: Enhancing Masked Autoencoders with Feature Mimicking

arXiv: https://arxiv.org/abs/2303.05475 · HTML: https://mlsanity.com/paper/2303.05475/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- Builds on 49 of its 59 references that are in the map.
- First posted on arXiv 9 Mar 2023.

## Builds on (most cited first, 30 of 49)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Mask R-CNN](https://mlsanity.com/paper/1703.06870/) (2017)
- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [Hierarchical Text-Conditional Image Generation with CLIP Latents](https://mlsanity.com/paper/2204.06125/) (2022)
- [Feature Pyramid Networks for Object Detection](https://mlsanity.com/paper/1612.03144/) (2016)
- [Zero-Shot Text-to-Image Generation](https://mlsanity.com/paper/2102.12092/) (2021)
- [BEiT: BERT Pre-Training of Image Transformers](https://mlsanity.com/paper/2106.08254/) (2021)
- [Unsupervised Feature Learning via Non-Parametric Instance-level Discrimination](https://mlsanity.com/paper/1805.01978/) (2018)
- [Decoupled Weight Decay Regularization](https://mlsanity.com/paper/1711.05101/) (2017)
- [An Empirical Study of Training Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.02057/) (2021)
- [DINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detection](https://mlsanity.com/paper/2203.03605/) (2022)
- [SimMIM: A Simple Framework for Masked Image Modeling](https://mlsanity.com/paper/2111.09886/) (2021)
- [data2vec: A General Framework for Self-supervised Learning in Speech, Vision and Language](https://mlsanity.com/paper/2202.03555/) (2022)
- [Exploring Plain Vision Transformer Backbones for Object Detection](https://mlsanity.com/paper/2203.16527/) (2022)
- [Masked Feature Prediction for Self-Supervised Visual Pre-Training](https://mlsanity.com/paper/2112.09133/) (2021)
- [Early Convolutions Help Transformers See Better](https://mlsanity.com/paper/2106.14881/) (2021)
- [CMT: Convolutional Neural Networks Meet Vision Transformers](https://mlsanity.com/paper/2107.06263/) (2021)
- [BEiT v2: Masked Image Modeling with Vector-Quantized Visual Tokenizers](https://mlsanity.com/paper/2208.06366/) (2022)
- [Context Autoencoder for Self-Supervised Representation Learning](https://mlsanity.com/paper/2202.03026/) (2022)
- [Point-M2AE: Multi-scale Masked Autoencoders for Hierarchical Point Cloud Pre-training](https://mlsanity.com/paper/2205.14401/) (2022)

## Built on by (most cited first, 12 of 12)

- [Not All Features Matter: Enhancing Few-shot CLIP with Adaptive Prior Refinement](https://mlsanity.com/paper/2304.01195/) (2023)
- [PiMAE: Point Cloud and Image Interactive Masked Autoencoders for 3D Object Detection](https://mlsanity.com/paper/2303.08129/) (2023)
- [Know Your Self-supervised Learning: A Survey on Image-based Generative and Discriminative Training](https://mlsanity.com/paper/2305.13689/) (2023)
- [Continual-MAE: Adaptive Distribution Masked Autoencoders for Continual Test-Time Adaptation](https://mlsanity.com/paper/2312.12480/) (2023)
- [Masked Angle-Aware Autoencoder for Remote Sensing Images](https://mlsanity.com/paper/2408.01946/) (2024)
- [One Leaf Reveals the Season: Occlusion-Based Contrastive Learning with Semantic-Aware Views for Efficient Visual Representation](https://mlsanity.com/paper/2411.09858/) (2024)
- [TIPSv2: Advancing Vision-Language Pretraining with Enhanced Patch-Text Alignment](https://mlsanity.com/paper/2604.12012/) (2026)
- [LoRA-TTT: Low-Rank Test-Time Training for Vision-Language Models](https://mlsanity.com/paper/2502.02069/) (2025)
- [Self-Supervised Skeleton-Based Action Representation Learning: A Benchmark and Beyond](https://mlsanity.com/paper/2406.02978/) (2024)
- [Enhancing Representations through Heterogeneous Self-Supervised Learning](https://mlsanity.com/paper/2310.05108/) (2023)
- [Image Recognition with Online Lightweight Vision Transformer: A Survey](https://mlsanity.com/paper/2505.03113/) (2025)
- [Orthogonal Projection Subspace to Aggregate Online Prior-knowledge for Continual Test-time Adaptation](https://mlsanity.com/paper/2506.19022/) (2025)

## Abstract (arXiv)

Masked Autoencoders (MAE) have been popular paradigms for large-scale vision representation pre-training. However, MAE solely reconstructs the low-level RGB signals after the decoder and lacks supervision upon high-level semantics for the encoder, thus suffering from sub-optimal learned representations and long pre-training epochs. To alleviate this, previous methods simply replace the pixel reconstruction targets of 75% masked tokens by encoded features from pre-trained image-image (DINO) or image-language (CLIP) contrastive learning. Different from those efforts, we propose to Mimic before Reconstruct for Masked Autoencoders, named as MR-MAE, which jointly learns high-level and low-level representations without interference during pre-training. For high-level semantics, MR-MAE employs a mimic loss over 25% visible tokens from the encoder to capture the pre-trained patterns encoded in CLIP and DINO. For low-level structures, we inherit the reconstruction loss in MAE to predict RGB pixel values for 75% masked tokens after the decoder. As MR-MAE applies high-level and low-level targets respectively at different partitions, the learning conflicts between them can be naturally overcome and contribute to superior visual representations for various downstream tasks. On ImageNet-1K, the MR-MAE base pre-trained for only 400 epochs achieves 85.8% top-1 accuracy after fine-tuning, surpassing the 1600-epoch MAE base by +2.2% and the previous state-of-the-art BEiT V2 base by +0.3%. Code and pre-trained models will be released at https://github.com/Alpha-VL/ConvMAE.