Papers with Code 아카이브

Efficient ViTs

태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.

하위 태스크: 0
평가표: 3
보존된 평가 기록: 82

Increasing the efficiency of ViTs without the modification of the architecture. (i.e., Key & Query Sparsification, Token pruning & merging)

코드

관련 논문과 평가 결과에 연결된 구현입니다. 출처에서 확인된 저자 구현에는 공식 코드 표시가 있습니다.

https://github.com/BWLONG/BeyondAttentiveTokens

공식 코드

Beyond Attentive Tokens: Incorporating Token Importance and Diversity for Efficient Vision Transformers

https://github.com/MonashAI/HVT

공식 코드

Scalable Vision Transformers with Hierarchical Pooling

https://github.com/VITA-Group/SViTE

공식 코드

Chasing Sparsity in Vision Transformers: An End-to-End Exploration

https://github.com/YifanXu74/Evo-ViT

공식 코드

Evo-ViT: Slow-Fast Token Evolution for Dynamic Vision Transformer

https://github.com/ackesnal/gtp-vit

공식 코드

GTP-ViT: Efficient Vision Transformers via Graph-based Token Propagation

https://github.com/adaptivetokensampling/ATS

공식 코드

Adaptive Token Sampling For Efficient Vision Transformers

https://github.com/changsn/STViT-R

공식 코드

Making Vision Transformers Efficient from A Token Sparsification View

https://github.com/cydia2018/as-vit

공식 코드

Adaptive Sparse ViT: Towards Learnable Adaptive Token Pruning by Fully Exploiting Self-Attention

https://github.com/facebookresearch/deit

공식 코드

Training data-efficient image transformers & distillation through attention

https://github.com/facebookresearch/tome

공식 코드

Token Merging: Your ViT But Faster

https://github.com/gatech-eic/castling-vit

공식 코드

Castling-ViT: Compressing Self-Attention via Switching Towards Linear-Angular Attention at Vision Transformer Inference

https://github.com/gatech-eic/shiftaddvit

공식 코드

ShiftAddViT: Mixture of Multiplication Primitives Towards Efficient Vision Transformer

https://github.com/megvii-research/tps-cvpr2023

공식 코드

Joint Token Pruning and Squeezing Towards More Aggressive Compression of Vision Transformers

https://github.com/mindspore-lab/models

공식 코드

PPT: Token Pruning and Pooling for Efficient Vision Transformers

https://github.com/mlvlab/mctf

공식 코드

Multi-criteria Token Fusion with One-step-ahead Attention for Efficient Vision Transformers

https://github.com/mxbonn/ltmp

공식 코드

Learned Thresholds Token Merging and Pruning for Vision Transformers

https://github.com/opengvlab/diffrate

공식 코드

DiffRate : Differentiable Compression Rate for Efficient Vision Transformers

https://github.com/peiyanflying/spvit

공식 코드

SPViT: Enabling Faster Vision Transformers via Soft Token Pruning

https://github.com/raoyongming/DynamicViT

공식 코드

DynamicViT: Efficient Vision Transformers with Dynamic Token Sparsification

https://github.com/shihuihong214/trio-vit

공식 코드

Trio-ViT: Post-Training Quantization and Acceleration for Softmax-Free Efficient Vision Transformer

https://github.com/siyi-wind/mdvit

공식 코드

MDViT: Multi-domain Vision Transformer for Small Medical Image Segmentation Datasets

https://github.com/xjwu1024/PPT

공식 코드

PPT: Token Pruning and Pooling for Efficient Vision Transformers

https://github.com/youweiliang/evit

공식 코드

Not All Patches are What You Need: Expediting Vision Transformers via Token Reorganizations

https://github.com/zhuang-group/spvit

공식 코드

Pruning Self-attentions into Convolutional Layers in Single Path

https://github.com/zihangJiang/TokenLabeling

공식 코드

All Tokens Matter: Token Labeling for Training Better Vision Transformers

https://github.com/zip-group/litv2

공식 코드

Fast Vision Transformers with HiLo Attention

https://github.com/zip-group/spvit

공식 코드

Pruning Self-attentions into Convolutional Layers in Single Path

https://github.com/ziplab/litv2

공식 코드

Fast Vision Transformers with HiLo Attention

https://github.com/ziplab/spvit

공식 코드

Pruning Self-attentions into Convolutional Layers in Single Path

출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.