English
Related papers

Related papers: Training-Free Acceleration of ViTs with Delayed Sp…

200 papers

The quadratic computational complexity to the number of tokens limits the practical applications of Vision Transformers (ViTs). Several works propose to prune redundant tokens to achieve efficient ViTs. However, these methods generally…

Computer Vision and Pattern Recognition · Computer Science 2023-03-31 Shuning Chang , Pichao Wang , Ming Lin , Fan Wang , David Junhao Zhang , Rong Jin , Mike Zheng Shou

Vision Transformers (ViTs) can learn strong image-level representations while their patch representations become less effective for dense prediction during prolonged training. We revisit this dense degradation phenomenon and argue that it…

Computer Vision and Pattern Recognition · Computer Science 2026-05-25 Linxiang Su

Vision Transformer (ViT) models have made breakthroughs in image embedding extraction, which provide state-of-the-art performance in tasks such as zero-shot image classification. However, the models suffer from a high computational burden.…

Computer Vision and Pattern Recognition · Computer Science 2025-06-03 Takahiro Naruko , Hiroaki Akutsu

Vision Transformer (ViT) has emerged as a prominent backbone for computer vision. For more efficient ViTs, recent works lessen the quadratic cost of the self-attention layer by pruning or fusing the redundant tokens. However, these works…

Computer Vision and Pattern Recognition · Computer Science 2024-04-02 Sanghyeok Lee , Joonmyung Choi , Hyunwoo J. Kim

Self-attention-based vision transformers (ViTs) have emerged as a highly competitive architecture in computer vision. Unlike convolutional neural networks (CNNs), ViTs are capable of global information sharing. With the development of…

Computer Vision and Pattern Recognition · Computer Science 2023-09-25 Zhenzhen Chu , Jiayu Chen , Cen Chen , Chengyu Wang , Ziheng Wu , Jun Huang , Weining Qian

Self-attention and transformers have been widely used in deep learning. Recent efforts have been devoted to incorporating transformer blocks into different neural architectures, including those with convolutions, leading to various visual…

Computer Vision and Pattern Recognition · Computer Science 2025-07-22 Yancheng Wang , Yingzhen Yang

Many modern ViT backbones adopt spatial architectural designs, such as window attention, decomposed relative positional embeddings in SAM, and RoPE in DINOv3. Such architectures impose new challenges on token reduction, as the vast majority…

Computer Vision and Pattern Recognition · Computer Science 2026-03-03 Wenyi Gong , Mieszko Lis

Vision Transformers can achieve high accuracy and strong generalization across various contexts, but their practical applicability on real-world robotic systems is limited due to their quadratic attention complexity. Recent works have…

Computer Vision and Pattern Recognition · Computer Science 2026-05-04 Fabio Montello , Ronja Güldenring , Lazaros Nalpantidis

We present a novel method that extends the self-attention mechanism of a vision transformer (ViT) for more accurate object detection across diverse datasets. ViTs show strong capability for image understanding tasks such as object…

Computer Vision and Pattern Recognition · Computer Science 2024-12-30 Tan Nguyen , Coy D. Heldermon , Corey Toler-Franklin

Vision transformers (ViTs) have emerged as a prevalent architecture for vision tasks owing to their impressive performance. However, when it comes to handling long token sequences, especially in dense prediction tasks that require…

Computer Vision and Pattern Recognition · Computer Science 2023-11-03 Jin Li , Yaoming Wang , Xiaopeng Zhang , Bowen Shi , Dongsheng Jiang , Chenglin Li , Wenrui Dai , Hongkai Xiong , Qi Tian

Vision transformers (ViT) usually extract features via forwarding all the tokens in the self-attention layers from top to toe. In this paper, we introduce dynamic token-pass vision transformers (DoViT) for semantic segmentation, which can…

Computer Vision and Pattern Recognition · Computer Science 2023-08-25 Yuang Liu , Qiang Zhou , Jing Wang , Fan Wang , Jun Wang , Wei Zhang

The adoption of Vision Transformers (ViTs) in resource-constrained applications necessitates improvements in inference throughput. To this end several token pruning and merging approaches have been proposed that improve efficiency by…

Computer Vision and Pattern Recognition · Computer Science 2024-12-03 Benjamin Bergner , Christoph Lippert , Aravindh Mahendran

Vision-Language-Action (VLA) models pretrained on large-scale multimodal datasets have emerged as powerful foundations for robotic perception and control. However, their massive scale, often billions of parameters, poses significant…

Robotics · Computer Science 2025-12-11 Yifan Ye , Jiaqi Ma , Jun Cen , Zhihe Lu

Vision Transformers (ViTs) have achieved state-of-the-art performance on various vision tasks. However, ViTs' self-attention module is still arguably a major bottleneck, limiting their achievable hardware efficiency. Meanwhile, existing…

Machine Learning · Computer Science 2025-03-04 Haoran You , Zhanyi Sun , Huihong Shi , Zhongzhi Yu , Yang Zhao , Yongan Zhang , Chaojian Li , Baopu Li , Yingyan Celine Lin

Token compression expedites the training and inference of Vision Transformers (ViTs) by reducing the number of the redundant tokens, e.g., pruning inattentive tokens or merging similar tokens. However, when applied to downstream tasks,…

Computer Vision and Pattern Recognition · Computer Science 2024-08-14 Shibo Jie , Yehui Tang , Jianyuan Guo , Zhi-Hong Deng , Kai Han , Yunhe Wang

In this paper, we present token labeling -- a new training objective for training high-performance vision transformers (ViTs). Different from the standard training objective of ViTs that computes the classification loss on an additional…

Computer Vision and Pattern Recognition · Computer Science 2021-06-10 Zihang Jiang , Qibin Hou , Li Yuan , Daquan Zhou , Yujun Shi , Xiaojie Jin , Anran Wang , Jiashi Feng

Vision Transformer (ViT) has achieved impressive results across various vision tasks, yet its high computational cost limits practical applications. Recent methods have aimed to reduce ViT's $O(n^2)$ complexity by pruning unimportant…

Computer Vision and Pattern Recognition · Computer Science 2025-07-17 Yi-Kuan Hsieh , Jun-Wei Hsieh , Xin Li , Yu-Ming Chang , Yu-Chee Tseng

Transformer-based models have driven significant advancements in Multimodal Large Language Models (MLLMs), yet their computational costs surge drastically when scaling resolution, training data, and model parameters. A key bottleneck stems…

Computer Vision and Pattern Recognition · Computer Science 2025-07-04 Weili Zeng , Ziyuan Huang , Kaixiang Ji , Yichao Yan

Video Transformers have become the prevalent solution for various video downstream tasks with superior expressive power and flexibility. However, these video transformers suffer from heavy computational costs induced by the massive number…

Computer Vision and Pattern Recognition · Computer Science 2024-04-02 Joonmyung Choi , Sanghyeok Lee , Jaewon Chu , Minhyuk Choi , Hyunwoo J. Kim

The success of VLMs often relies on the dynamic high-resolution schema that adaptively augments the input images to multiple crops, so that the details of the images can be retained. However, such approaches result in a large number of…

Computer Vision and Pattern Recognition · Computer Science 2025-02-04 Jiayi Han , Liang Du , Yiwen Wu , Xiangguo Zhou , Hongwei Du , Weibo Zheng