English
Related papers

Related papers: Effective Theory of Transformers at Initialization

200 papers

Despite the growing use of transformer models in computer vision, a mechanistic understanding of these networks is still needed. This work introduces a method to reverse-engineer Vision Transformers trained to solve image classification…

Computer Vision and Pattern Recognition · Computer Science 2023-10-31 Martina G. Vilas , Timothy Schaumlöffel , Gemma Roig

In-context learning (ICL) of large language models has proven to be a surprisingly effective method of learning a new task from only a few demonstrative examples. In this paper, we study the efficacy of ICL from the viewpoint of statistical…

Machine Learning · Statistics 2024-10-03 Juno Kim , Tai Nakamaki , Taiji Suzuki

Inspired by the success of Transformer-based models in natural language processing, this paper investigates their potential as foundation models for network traffic analysis. We propose a unified pre-training and fine-tuning pipeline for…

Networking and Internet Architecture · Computer Science 2026-02-09 Samara Mayhoub , Chuan Heng Foh , Mahdi Boloursaz Mashhadi , Mohammad Shojafar , Rahim Tafazolli

Transformer models cannot easily scale to long sequences due to their O(N^2) time and space complexity. This has led to Transformer variants seeking to lower computational complexity, such as Longformer and Performer. While such models have…

Computation and Language · Computer Science 2024-12-10 Guanghui Qin , Yukun Feng , Benjamin Van Durme

This document is a follow-up to our previous paper dedicated to a vectorized derivation of backpropagation in CNNs. Following the same principles and notations already put in place there, we now focus on transformer-based…

Machine Learning · Computer Science 2025-12-30 Laurent Boué

Conventional wisdom suggests that pre-training Vision Transformers (ViT) improves downstream performance by learning useful representations. Is this actually true? We investigate this question and find that the features and representations…

Machine Learning · Computer Science 2024-11-15 Alexander C. Li , Yuandong Tian , Beidi Chen , Deepak Pathak , Xinlei Chen

Deep learning frameworks have become powerful tools for approaching scientific problems such as turbulent flow, which has wide-ranging applications. In practice, however, existing scientific machine learning approaches have difficulty…

Machine Learning · Computer Science 2024-07-25 Jakin Ng , Yongji Wang , Ching-Yao Lai

Large Language Models (LLMs) are trained with next-token prediction, implemented in autoregressive Transformers via causal masking for parallelism. This creates a subtle misalignment: residual connections tie activations to the current…

Computation and Language · Computer Science 2026-03-03 Jonathan Lys , Vincent Gripon , Bastien Pasdeloup , Axel Marmoret , Lukas Mauch , Fabien Cardinaux , Ghouthi Boukli Hacene

Learning deeper convolutional neural networks becomes a tendency in recent years. However, many empirical evidences suggest that performance improvement cannot be gained by simply stacking more layers. In this paper, we consider the issue…

Computer Vision and Pattern Recognition · Computer Science 2016-04-05 Li Shen , Zhouchen Lin , Qingming Huang

Weight initialization governs signal propagation and gradient flow at the start of training. This paper offers a theory-grounded and empirically validated study across two regimes: compact ReLU multilayer perceptrons and GPT-2-style…

Machine Learning · Computer Science 2025-10-13 Yankun Han

This paper proposes Omnidirectional Representations from Transformers (OmniNet). In OmniNet, instead of maintaining a strictly horizontal receptive field, each token is allowed to attend to all tokens in the entire network. This process can…

Computer Vision and Pattern Recognition · Computer Science 2021-03-02 Yi Tay , Mostafa Dehghani , Vamsi Aribandi , Jai Gupta , Philip Pham , Zhen Qin , Dara Bahri , Da-Cheng Juan , Donald Metzler

The lack of theoretical results for Layer Normalization and feedforward Hessians has left a gap in the study of Transformer optimization landscapes. We address this by deriving explicit second-order expressions for these components, thereby…

Machine Learning · Computer Science 2025-10-21 Egor Petrov , Nikita Kiselev , Vladislav Meshkov , Andrey Grabovoy

In this report, we combine the idea of Wide ResNets and transfer learning to optimize the architecture of deep neural networks. The first improvement of the architecture is the use of all layers as information source for the last layer.…

Machine Learning · Computer Science 2022-06-22 Wolfgang Fuhl

The Transformer is a fully attention-based alternative to recurrent networks that has achieved state-of-the-art results across a range of NLP tasks. In this paper, we analyze the structure of attention in a Transformer language model, the…

Computation and Language · Computer Science 2019-06-20 Jesse Vig , Yonatan Belinkov

Fine-tuning pre-trained transformers is a powerful technique for enhancing the performance of base models on specific tasks. From early applications in models like BERT to fine-tuning Large Language Models (LLMs), this approach has been…

Computation and Language · Computer Science 2025-02-25 Suneel Nadipalli

Large Language Models (LLMs) have achieved impressive performance across diverse tasks but continue to struggle with learning transitive relations, a cornerstone for complex planning. To address this issue, we investigate the Multi-Token…

Artificial Intelligence · Computer Science 2025-09-30 Qimin Zhong , Hao Liao , Siwei Wang , Mingyang Zhou , Xiaoqun Wu , Rui Mao , Wei Chen

Large language models based on transformer architectures have become integral to state-of-the-art natural language processing applications. However, their training remains computationally expensive and exhibits instabilities, some of which…

Numerical Analysis · Mathematics 2025-03-14 Stanislav Budzinskiy , Wenyi Fang , Longbin Zeng , Philipp Petersen

The concept of a recently proposed Forward-Forward learning algorithm for fully connected artificial neural networks is applied to a single multi output perceptron for classification. The parameters of the system are trained with respect to…

Machine Learning · Computer Science 2023-04-07 K. Fredrik Karlsson

Preceptron model updating with back propagation has become the routine of deep learning. Continuous feed forward procedure is required in order for backward propagate to function properly. Doubting the underlying physical interpretation on…

Signal Processing · Electrical Eng. & Systems 2020-10-19 Shirui Tang

Recently, end-to-end sequence-to-sequence models for speech recognition have gained significant interest in the research community. While previous architecture choices revolve around time-delay neural networks (TDNN) and long short-term…

Computation and Language · Computer Science 2019-05-06 Ngoc-Quan Pham , Thai-Son Nguyen , Jan Niehues , Markus Müller , Sebastian Stüker , Alexander Waibel
‹ Prev 1 3 4 5 6 7 10 Next ›