English
Related papers

Related papers: Softmax-free Linear Transformers

200 papers

Vision transformers have demonstrated remarkable success in classification by leveraging global self-attention to capture long-range dependencies. However, this same mechanism can obscure fine-grained spatial details crucial for tasks such…

Computer Vision and Pattern Recognition · Computer Science 2026-03-06 Sina Hajimiri , Farzad Beizaee , Fereshteh Shakeri , Christian Desrosiers , Ismail Ben Ayed , Jose Dolz

Despite the recent success in many applications, the high computational requirements of vision transformers limit their use in resource-constrained settings. While many existing methods improve the quadratic complexity of attention, in most…

Computer Vision and Pattern Recognition · Computer Science 2023-02-28 Dmitrii Marin , Jen-Hao Rick Chang , Anurag Ranjan , Anish Prabhu , Mohammad Rastegari , Oncel Tuzel

Transformers have recently demonstrated strong performance in computer vision, with Vision Transformers (ViTs) leveraging self-attention to capture both low-level and high-level image features. However, standard ViTs remain computationally…

Computer Vision and Pattern Recognition · Computer Science 2026-01-08 Ali El Bellaj , Mohammed-Amine Cheddadi , Rhassan Berber

The attention mechanism is the computational core of modern Transformer architectures, but its quadratic complexity in the input sequence length is the bottleneck for large-scale inference. This has motivated a rapidly growing body of work…

Over the past few years, vision transformers (ViTs) have consistently demonstrated remarkable performance across various visual recognition tasks. However, attempts to enhance their robustness have yielded limited success, mainly focusing…

Computer Vision and Pattern Recognition · Computer Science 2024-10-01 Nick Nikzad , Yi Liao , Yongsheng Gao , Jun Zhou

Vision Transformer (ViT) is emerging as the state-of-the-art architecture for image recognition. While recent studies suggest that ViTs are more robust than their convolutional counterparts, our experiments find that ViTs trained on…

Computer Vision and Pattern Recognition · Computer Science 2022-04-05 Chengzhi Mao , Lu Jiang , Mostafa Dehghani , Carl Vondrick , Rahul Sukthankar , Irfan Essa

Generic re-usable pre-trained image representation encoders have become a standard component of methods for many computer vision tasks. As visual representations for robots however, their utility has been limited, leading to a recent wave…

Computer Vision and Pattern Recognition · Computer Science 2024-05-28 Jianing Qian , Anastasios Panagopoulos , Dinesh Jayaraman

This paper investigates the limitations of the normalization in attention mechanisms. We begin with a theoretical framework that enables the identification of the model's selective ability and the geometric separation involved in token…

Machine Learning · Computer Science 2025-10-21 Timur Mudarisov , Mikhail Burtsev , Tatiana Petrova , Radu State

In recent years, vision transformers (ViTs) have emerged as powerful and promising techniques for computer vision tasks such as image classification, object detection, and segmentation. Unlike convolutional neural networks (CNNs), which…

Computer Vision and Pattern Recognition · Computer Science 2025-05-20 Shaibal Saha , Lanyu Xu

While linear attention reduces the quadratic complexity of standard Transformers to linear time, it often lags behind in expressivity due to the removal of softmax normalization. This omission eliminates \emph{global competition}, a…

Machine Learning · Computer Science 2026-02-03 Mingwei Xu , Xuan Lin , Xinnan Guo , Wanqing Xu , Wanyun Cui

The softmax attention mechanism has emerged as a noteworthy development in the field of Artificial Intelligence research, building on the successes of Transformer-based architectures. However, their ever increasing sizes necessitate ever…

Computation and Language · Computer Science 2023-05-19 Luca Herranz-Celotti , Ermal Rrapaj

Convolutional Neural Networks (CNNs), architectures consisting of convolutional layers, have been the standard choice in vision tasks. Recent studies have shown that Vision Transformers (VTs), architectures based on self-attention modules,…

Computer Vision and Pattern Recognition · Computer Science 2022-01-24 Kishaan Jeeveswaran , Senthilkumar Kathiresan , Arnav Varma , Omar Magdy , Bahram Zonooz , Elahe Arani

Transformers have recently gained increasing attention in computer vision. However, existing studies mostly use Transformers for feature representation learning, e.g. for image classification and dense predictions, and the generalizability…

Computer Vision and Pattern Recognition · Computer Science 2021-12-08 Shengcai Liao , Ling Shao

Vision Transformers (ViTs) have redefined image classification by leveraging self-attention to capture complex patterns and long-range dependencies between image patches. However, a key challenge for ViTs is efficiently incorporating…

Computer Vision and Pattern Recognition · Computer Science 2025-01-09 Shravan Venkatraman , Jaskaran Singh Walia , Joe Dhanith P R

Convolutional architectures have proven extremely successful for vision tasks. Their hard inductive biases enable sample-efficient learning, but come at the cost of a potentially lower performance ceiling. Vision Transformers (ViTs) rely on…

Computer Vision and Pattern Recognition · Computer Science 2022-12-07 Stéphane d'Ascoli , Hugo Touvron , Matthew Leavitt , Ari Morcos , Giulio Biroli , Levent Sagun

Vision Transformers (ViTs) have achieved state-of-the-art performance in image classification, yet their attention mechanisms often remain opaque and exhibit dense, non-structured behaviors. In this work, we adapt our previously proposed…

Computer Vision and Pattern Recognition · Computer Science 2026-02-12 Vasileios Arampatzakis , George Pavlidis , Nikolaos Mitianoudis , Nikos Papamarkos

Vision transformers (ViTs) are usually considered to be less light-weight than convolutional neural networks (CNNs) due to the lack of inductive bias. Recent works thus resort to convolutions as a plug-and-play module and embed them in…

Computer Vision and Pattern Recognition · Computer Science 2022-07-13 Tao Huang , Lang Huang , Shan You , Fei Wang , Chen Qian , Chang Xu

Vision Transformers and their variants have achieved remarkable success in diverse visual perception tasks. Despite their effectiveness, they suffer from two significant limitations. First, the quadratic computational complexity of…

Computer Vision and Pattern Recognition · Computer Science 2026-02-16 Ali K. Rahimian , Manish K. Govind , Subhajit Maity , Dominick Reilly , Christian Kümmerle , Srijan Das , Aritra Dutta

The quadratic computational complexity of softmax transformers has become a bottleneck in long-context scenarios. In contrast, linear attention model families provide a promising direction towards a more efficient sequential model. These…

Computation and Language · Computer Science 2026-02-04 Difan Deng , Andreas Bentzen Winje , Lukas Fehring , Marius Lindauer

In classification tasks, softmax functions are ubiquitously used as output activations to produce predictive probabilities. Such outputs only capture aleatoric uncertainty. To capture epistemic uncertainty, approximate Gaussian inference…

Machine Learning · Computer Science 2026-02-12 Bálint Mucsányi , Nathaël Da Costa , Philipp Hennig