Research map / Detection and segmentation / CNN features
Evolved Hierarchical Masking for Self-Supervised Learning
Zhanzhou Feng, Shiliang Steven Zhang
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 changed | Where | Measured | Effect |
|---|---|---|---|
| evolve masking depth during training instead of fixed shallow depth (ablation) | evolved hierarchical masking · ImageNet-1K top-1 (ViT-S) | top-1 accuracy: 79.05 -> 80.41 | Improved |
| evolve masking depth instead of fixed shallow depth (ablation) | evolved hierarchical masking · ADE20K (ViT-S) | mIoU: 39.86 -> 41.37 | Improved |
| evolve masking depth instead of fixed deep depth (ablation) | evolved hierarchical masking · Oxford retrieval (ViT-S) | mAP: 55.07 -> 57.41 | Improved |
| evolve masking depth instead of fixed shallow depth (ablation) | evolved hierarchical masking · Oxford retrieval (ViT-S) | mAP: 57.11 -> 57.41 | No clear effect |
| use hierarchical shallow masking instead of grid-like masking (ablation) | hierarchical masking · ImageNet-1K top-1 (ViT-S) | top-1 accuracy: 73.16 -> 79.05 | Improved |
| use hierarchical deep masking instead of large block masking (ablation) | hierarchical masking · ImageNet-1K top-1 (ViT-S) | top-1 accuracy: 77.89 -> 78.76 | Improved |
| use hierarchical masking instead of random masking at mask ratio 0.25 (ablation) | MAE · ImageNet-1K top-1 (ViT-S) | top-1 accuracy: 77.19 -> 78.96 | Improved |
| use hierarchical masking instead of random masking at mask ratio 0.75 (ablation) | MAE · ADE20K (ViT-S) | mIoU: 39.24 -> 40.59 | Improved |
| add evolved hierarchical masking | MAE · ImageNet-1K top-1 (200 ep pretrain) | top-1 accuracy: 79.20 -> 80.06 | Improved |
| add evolved hierarchical masking | SimMIM · ImageNet-1K top-1 (200 ep pretrain) | top-1 accuracy: 78.51 -> 79.21 | Improved |
| add evolved hierarchical masking | MAE · ADE20K (200 ep pretrain) | mIoU: 38.81 -> 40.23 | Improved |
| extend pretraining from 100 to 2400 epochs | evolved hierarchical masking · ImageNet-1K top-1 (ViT-S) | top-1 accuracy: 78.29 -> 81.55 | Improved |
Closest papers by meaning
- Evolved Part Masking for Self-Supervised Learning (CVPR 2023 · similarity 0.94)
- Progressive Mask Distillation for Self-supervised Video Representation (CVPR 2026 · similarity 0.89)
- From Prototypes to General Distributions: An Efficient Curriculum for Masked Image Modeling (CVPR 2025 · similarity 0.89)
- Good Helper Is around You: Attention-Driven Masked Image Modeling (AAAI 2023 · similarity 0.88)
- Masked Scene Modeling: Narrowing the Gap Between Supervised and Self-Supervised Learning in 3D Scene Understanding (CVPR 2025 · similarity 0.87)
- Self-Guided Masked Autoencoder (NeurIPS 2024 · similarity 0.87)
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.
