神经架构搜索论文整理(自用，不全)

1. 神经架构搜索

1.1 Transformer的搜索

- Training-free Transformer Architecture Search 2022CVPR oral

paper: https://arxiv.org/abs/2203.12217

code: https://github.com/decemberzhou/TF_TAS (06.09暂未公布)

1.2 CNN的搜索

- Global Convergence of MAML and Theory-Inspired Neural Architecture Search for Few-Shot Learning 2022CVPR

paper: https://arxiv.org/abs/2203.09137

code: https://github.com/YiteWang/MetaNTK-NAS

- β-DARTS: Beta-Decay Regularization for Differentiable Architecture Search 2022CVPR

paper: https://arxiv.org/abs/2203.01665

code: https://github.com/Sunshine-Ye/Beta-DARTS

- HourNAS: Extremely Fast Neural Architecture 2021CVPR

paper: https://arxiv.org/pdf/2005.14446.pdf

- ReNAS: Relativistic Evaluation of Neural Architecture Search 2021CVPR

paper: https://arxiv.org/pdf/1910.01523.pdf

code:

- AttentiveNAS: Improving Neural Architecture Search via Attentive 2021CVPR

paper:https://arxiv.org/pdf/2011.09011.pdf

code:

- OPANAS: One-Shot Path Aggregation Network Architecture Search for Object 2021CVPR

paper: https://arxiv.org/abs/2103.04507

code:

- Contrastive Neural Architecture Search with Neural Architecture Comparators 2021CVPR

paper: https://arxiv.org/abs/2103.05471

code:

- Searching by Generating: Flexible and Efficient One-Shot NAS with Architecture  Generator 2021CVPR

paper: https://arxiv.org/abs/2103.07289

code:

- Prioritized Architecture Sampling with Monto-Carlo Tree Search 2021CVPR

paper: https://arxiv.org/pdf/2103.11922.pdf

code:

- Dynamic Slimmable Network 2021CVPR

paper: https://arxiv.org/abs/2103.13258

code:

2. Transformer网络架构设计

- Mobile-Former: Bridging MobileNet and Transformer 2022CVPR

paper: https://arxiv.org/abs/2108.05895

code: https://github.com/AAboys/MobileFormer (06.09暂未公布)

- Delving Deep into the Generalization of Vision Transformers under Distribution Shifts 2022CVPR

paper: https://arxiv.org/abs/2106.07617

code:GitHub - Phoenix1153/ViT_OOD_generalization

- NomMer: Nominate Synergistic Context in Vision Transformer for Visual Recognition 2022CVPR

paper: https://arxiv.org/abs/2111.12994

code: https://github.com/TencentYoutuResearch/VisualRecognition-NomMer

- Attribute Surrogates Learning and Spectral Tokens Pooling in Transformers for Few-shot Learning 2022CVPR

paper: https://arxiv.org/abs/2203.09064

code: GitHub - StomachCold/HCTransformers

- Bootstrapping ViTs: Towards Liberating Vision Transformers from Pre-training 2022CVPR

paper: https://arxiv.org/abs/2112.03552

code:https://github.com/zhfeing/Bootstrapping-ViTs-pytorch

- Transformer Interpretability Beyond Attention Visualization 2021CVPR

code: https://arxiv.org/pdf/2012.09838.pdf

paper: https://github.com/hila-chefer/Transformer-Explainability

- UP-DETR: Unsupervised Pre-training for Object Detection with Transformers 2021CVPR

paper: https://arxiv.org/pdf/2011.09094.pdf

code:https://github.com/dddzg/up-detr

- Pre-Trained Image Processing Transformer 2021CVPR

paper: https://arxiv.org/pdf/2012.00364.pdf

code: