English
Related papers

Related papers: Attention-Only Transformers via Unrolled Subspace …

200 papers

Attention layers, as commonly used in transformers, form the backbone of modern deep learning, yet there is no mathematical description of their benefits and deficiencies as compared with other architectures. In this work we establish both…

Machine Learning · Computer Science 2023-11-17 Clayton Sanford , Daniel Hsu , Matus Telgarsky

The Transformer architecture has revolutionized the field of sequence modeling and underpins the recent breakthroughs in large language models (LLMs). However, a comprehensive mathematical theory that explains its structure and operations…

Machine Learning · Computer Science 2026-04-14 Xue-Cheng Tai , Hao Liu , Lingfeng Li , Raymond H. Chan

Neural networks have greatly boosted performance in computer vision by learning powerful representations of input data. The drawback of end-to-end training for maximal overall performance are black-box models whose hidden representations…

Computer Vision and Pattern Recognition · Computer Science 2020-04-29 Patrick Esser , Robin Rombach , Björn Ommer

Learned denoisers play a fundamental role in various signal generation (e.g., diffusion models) and reconstruction (e.g., compressed sensing) architectures, whose success derives from their ability to leverage low-dimensional structure in…

Machine Learning · Computer Science 2025-08-14 Shiyu Wang , Mariam Avagyan , Yihan Shen , Arnaud Lamy , Tingran Wang , Szabolcs Márka , Zsuzsa Márka , John Wright

Transformer-based architectures are the most used architectures in many deep learning fields like Natural Language Processing, Computer Vision or Speech processing. It may encourage the direct use of Transformers in the constrained tasks,…

Audio and Speech Processing · Electrical Eng. & Systems 2026-01-29 Youness Dkhissi , Valentin Vielzeuf , Elys Allesiardo , Anthony Larcher

We study how a one-layer attention-only transformer develops relevant structures while learning to sort lists of numbers. At the end of training, the model organizes its attention heads in two main modes that we refer to as…

Machine Learning · Computer Science 2025-02-03 Einar Urdshals , Jasmina Urdshals

The quadratic cost of attention in transformers motivated the development of efficient approaches: namely sparse and sliding window attention, convolutions and linear attention. Although these approaches result in impressive reductions in…

Machine Learning · Computer Science 2025-11-10 Jatin Prakash , Aahlad Puli , Rajesh Ranganath

Transformers, adapted from natural language processing, are emerging as a leading approach for graph representation learning. Contemporary graph transformers often treat nodes or edges as separate tokens. This approach leads to…

Machine Learning · Computer Science 2023-10-04 Zihan Pengmei , Zimu Li , Chih-chan Tien , Risi Kondor , Aaron R. Dinner

We introduce a constrained optimization framework for training transformers that behave like optimization descent algorithms. Specifically, we enforce layerwise descent constraints on the objective function and replace standard empirical…

Machine Learning · Computer Science 2026-01-27 Javier Porras-Valenzuela , Samar Hadou , Alejandro Ribeiro

Transformer architecture has been very successful long runner in the field of Deep Learning (DL) and Large Language Models (LLM) because of its powerful attention-based learning and parallel-natured architecture. As the models grow gigantic…

Machine Learning · Computer Science 2026-01-21 Phani Kumar , Nyshadham , Jyothendra Varma , Polisetty V R K , Aditya Rathore

Attention mechanisms have achieved significant empirical success in multiple fields, but their underlying optimization objectives remain unclear yet. Moreover, the quadratic complexity of self-attention has become increasingly prohibitive.…

Machine Learning · Computer Science 2025-11-06 Qishuai Wen , Zhiyuan Huang , Chun-Guang Li

Blind-spot networks (BSN) have been prevalent neural architectures in self-supervised image denoising (SSID). However, most existing BSNs are conducted with convolution layers. Although transformers have shown the potential to overcome the…

Computer Vision and Pattern Recognition · Computer Science 2025-08-20 Junyi Li , Zhilu Zhang , Wangmeng Zuo

Traditional supervised denoising networks learn network weights through "black box" (pixel-oriented) training, which requires clean training labels. The uninterpretability nature of such denoising networks in addition to the requirement for…

Geophysics · Physics 2023-08-08 Sixiu Liu , Shijun Cheng , Tariq Alkhalifah

This work introduces a Transformer-based image compression system. It has the flexibility to switch between the standard image reconstruction and the denoising reconstruction from a single compressed bitstream. Instead of training separate…

Image and Video Processing · Electrical Eng. & Systems 2024-02-21 Yi-Hsin Chen , Kuan-Wei Ho , Shiau-Rung Tsai , Guan-Hsun Lin , Alessandro Gnutti , Wen-Hsiao Peng , Riccardo Leonardi

Transformers have emerged as a powerful neural network architecture capable of tackling a wide range of learning tasks. In this work, we provide a theoretical analysis of their ability to automatically extract structure from data in an…

Machine Learning · Statistics 2025-10-29 Rodrigo Maulen-Soto , Pierre Marion , Claire Boyer

Fine-tuned transformer models have shown superior performances in many natural language tasks. However, the large model size prohibits deploying high-performance transformer models on resource-constrained devices. This paper proposes a…

Computation and Language · Computer Science 2024-10-01 Zi Yang , Samridhi Choudhary , Siegfried Kunzmann , Zheng Zhang

Encoder transformer models compress information from all tokens in a sequence into a single [CLS] token to represent global context. This approach risks diluting fine-grained or hierarchical features, leading to information loss in…

Computation and Language · Computer Science 2025-09-23 Asif Shahriar , Rifat Shahriyar , M Saifur Rahman

Transformer-based QA models use input-wide self-attention -- i.e. across both the question and the input passage -- at all layers, causing them to be slow and memory-intensive. It turns out that we can get by without input-wide…

Computation and Language · Computer Science 2020-05-05 Qingqing Cao , Harsh Trivedi , Aruna Balasubramanian , Niranjan Balasubramanian

The reasoning abilities of Large Language Models (LLMs) can be improved by structurally denoising their weights, yet existing techniques primarily focus on denoising the feed-forward network (FFN) of the transformer block, and can not…

Computation and Language · Computer Science 2025-05-16 Yuxuan Gu , Wuyang Zhou , Giorgos Iacovides , Danilo Mandic

Mechanical vibration signal denoising is a pivotal task in various industrial applications, including system health monitoring and failure prediction. This paper introduces a novel deep learning transformer-based architecture specifically…

Systems and Control · Electrical Eng. & Systems 2023-08-07 Han Chen , Yang Yu , Pengtao Li