中文
相关论文

相关论文: Bridging The Gaps Between Token Pruning and Full P…

200 篇论文

Vision Transformers (ViTs) have achieved remarkable success in various computer vision tasks. However, ViTs have a huge computational cost due to their inherent reliance on multi-head self-attention (MHSA), prompting efforts to accelerate…

计算机视觉与模式识别 · 计算机科学 2024-12-24 Seungdong Yoa , Seungjun Lee , Hyeseung Cho , Bumsoo Kim , Woohyung Lim

Vision Large Language Models (VLLMs) incur high computational costs due to their reliance on hundreds of visual tokens to represent images. While token pruning offers a promising solution for accelerating inference, this paper, however,…

计算机视觉与模式识别 · 计算机科学 2026-03-10 Yahong Wang , Juncheng Wu , Zhangkai Ni , Longzhen Yang , Yihang Liu , Chengmei Yang , Ying Wen , Lianghua He , Xianfeng Tang , Hui Liu , Yuyin Zhou

Prompt tuning and adapter tuning have shown great potential in transferring pre-trained vision-language models (VLMs) to various downstream tasks. In this work, we design a new type of tuning method, termed as regularized mask tuning, which…

计算机视觉与模式识别 · 计算机科学 2023-08-08 Kecheng Zheng , Wei Wu , Ruili Feng , Kai Zhu , Jiawei Liu , Deli Zhao , Zheng-Jun Zha , Wei Chen , Yujun Shen

Parameter-efficient fine-tuning (PEFT) has emerged as a popular solution for adapting pre-trained Vision Transformer (ViT) models to downstream applications by updating only a small subset of parameters. While current PEFT methods have…

计算机视觉与模式识别 · 计算机科学 2025-12-19 Ting Liu , Xuyang Liu , Liangtao Shi , Zunnan Xu , Yue Hu , Siteng Huang , Yi Xin , Bineng Zhong , Donglin Wang

The remarkable performance of Vision Transformers (ViTs) typically requires an extremely large training cost. Existing methods have attempted to accelerate the training of ViTs, yet typically disregard method universality with accuracy…

机器学习 · 计算机科学 2024-04-02 Wenxuan Huang , Yunhang Shen , Jiao Xie , Baochang Zhang , Gaoqi He , Ke Li , Xing Sun , Shaohui 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,…

计算机视觉与模式识别 · 计算机科学 2024-08-14 Shibo Jie , Yehui Tang , Jianyuan Guo , Zhi-Hong Deng , Kai Han , Yunhe Wang

Vision transformers have demonstrated remarkable success in a wide range of computer vision tasks over the last years. However, their high computational costs remain a significant barrier to their practical deployment. In particular, the…

计算机视觉与模式识别 · 计算机科学 2023-08-21 Maxim Bonnaerens , Joni Dambre

We propose an efficient approach to train large diffusion models with masked transformers. While masked transformers have been extensively explored for representation learning, their application to generative learning is less explored in…

计算机视觉与模式识别 · 计算机科学 2024-03-06 Hongkai Zheng , Weili Nie , Arash Vahdat , Anima Anandkumar

Visual token pruning is a widely used strategy for efficient inference in multimodal large language models (MLLMs), but existing work mainly evaluates it with task accuracy. In this paper, we study how visual token pruning affects model…

计算机视觉与模式识别 · 计算机科学 2026-04-15 Kaizhen Tan

Vision Transformers (ViTs) deliver state-of-the-art accuracy but their quadratic attention cost and redundant computations severely hinder deployment on latency and resource-constrained platforms. Existing pruning approaches treat either…

计算机视觉与模式识别 · 计算机科学 2025-12-24 Mohammad Helal Uddin , Liam Seymour , Sabur Baidya

Vision transformers (ViTs) have gained popularity recently. Even without customized image operators such as convolutions, ViTs can yield competitive performance when properly trained on massive data. However, the computational overhead of…

机器学习 · 计算机科学 2022-03-17 Shixing Yu , Tianlong Chen , Jiayi Shen , Huan Yuan , Jianchao Tan , Sen Yang , Ji Liu , Zhangyang Wang

Changes of camera perspective are a common obstacle in driver monitoring. While deep learning and pretrained foundation models show strong potential for improved generalization via lightweight adaptation of the final layers ('probing'),…

计算机视觉与模式识别 · 计算机科学 2025-08-26 Thinesh Thiyakesan Ponbagavathi , Kunyu Peng , Alina Roitberg

Learning a versatile language-image model is computationally prohibitive under a limited computing budget. This paper delves into the \emph{efficient language-image pre-training}, an area that has received relatively little attention…

计算机视觉与模式识别 · 计算机科学 2026-01-27 Yangyang Guo , Haoyu Zhang , Yongkang Wong , Liqiang Nie , Mohan Kankanhalli

Prompt learning has achieved great success in efficiently exploiting large-scale pre-trained models in natural language processing (NLP). It reformulates the downstream tasks as the generative pre-training ones to achieve consistency, thus…

计算机视觉与模式识别 · 计算机科学 2023-12-18 Ning Liao , Bowen Shi , Xiaopeng Zhang , Min Cao , Junchi Yan , Qi Tian

Masked Autoencoding (MAE) has emerged as an effective approach for pre-training representations across multiple domains. In contrast to discrete tokens in natural languages, the input for image MAE is continuous and subject to additional…

计算机视觉与模式识别 · 计算机科学 2022-10-14 Ronghang Hu , Shoubhik Debnath , Saining Xie , Xinlei Chen

Learning with large-scale unlabeled data has become a powerful tool for pre-training Visual Transformers (VTs). However, prior works tend to overlook that, in real-world scenarios, the input data may be corrupted and unreliable.…

计算机视觉与模式识别 · 计算机科学 2023-04-13 Tianjiao Li , Lin Geng Foo , Ping Hu , Xindi Shang , Hossein Rahmani , Zehuan Yuan , Jun Liu

Pre-training & fine-tuning can enhance the transferring efficiency and performance in visual tasks. Recent delta-tuning methods provide more options for visual classification tasks. Despite their success, existing visual delta-tuning art…

计算机视觉与模式识别 · 计算机科学 2024-08-28 Dongshuo Yin , Leiyi Hu , Bin Li , Youqun Zhang , Xue Yang

Training visual embeddings with labeled data supervision has been the de facto setup for representation learning in computer vision. Inspired by recent success of adopting masked image modeling (MIM) in self-supervised representation…

计算机视觉与模式识别 · 计算机科学 2023-12-05 Kaifeng Chen , Daniel Salz , Huiwen Chang , Kihyuk Sohn , Dilip Krishnan , Mojtaba Seyedhosseini

Vision Transformers (ViTs) often degrade under distribution shifts because they rely on spurious correlations, such as background cues, rather than semantically meaningful features. Existing regularization methods, typically relying on…

计算机视觉与模式识别 · 计算机科学 2026-03-17 Yehonatan Elisha , Oren Barkan , Noam Koenigstein

Inspired by the masked language modeling (MLM) in natural language processing tasks, the masked image modeling (MIM) has been recognized as a strong self-supervised pre-training method in computer vision. However, the high random mask ratio…

计算机视觉与模式识别 · 计算机科学 2024-06-04 Zhaowen Li , Yousong Zhu , Zhiyang Chen , Wei Li , Chaoyang Zhao , Rui Zhao , Ming Tang , Jinqiao Wang