English
Related papers

Related papers: Two Heads Are Better than One: Simulating Large Tr…

200 papers

Most neural networks utilize the same amount of compute for every example independent of the inherent complexity of the input. Further, methods that adapt the amount of computation to the example focus on finding a fixed inference-time…

Machine Learning · Computer Science 2020-04-17 Ankur Bapna , Naveen Arivazhagan , Orhan Firat

The attention mechanisms of transformers effectively extract pertinent information from the input sequence. However, the quadratic complexity of self-attention w.r.t the sequence length incurs heavy computational and memory burdens,…

Hardware Architecture · Computer Science 2022-06-30 Guan Shen , Jieru Zhao , Quan Chen , Jingwen Leng , Chao Li , Minyi Guo

Which transformer scaling regimes are able to perfectly solve different classes of algorithmic problems? While tremendous empirical advances have been attained by transformer-based neural networks, a theoretical understanding of their…

Transformer models have demonstrated high accuracy in numerous applications but have high complexity and lack sequential processing capability making them ill-suited for many streaming applications at the edge where devices are heavily…

Neural and Evolutionary Computing · Computer Science 2024-02-08 Zeyu Liu , Gourav Datta , Anni Li , Peter Anthony Beerel

With the evolution of large language models, traditional Transformer models become computationally demanding for lengthy sequences due to the quadratic growth in computation with respect to the sequence length. Mamba, emerging as a…

Machine Learning · Computer Science 2024-08-22 Haoran Xu , Ziqian Liu , Rong Fu , Zhongling Su , Zerui Wang , Zheng Cai , Zhilin Pei , Xingcheng Zhang

Foundational models based on the transformer architecture are currently the state-of-the-art in general language modeling, as well as in scientific areas such as material science and climate. However, training and deploying these models is…

Machine Learning · Computer Science 2025-10-16 Adarsha Balaji , Sandeep Madireddy , Prasanna Balaprakash

Transformers have emerged as the cornerstone of state-of-the-art natural language processing models, showcasing exceptional performance across a wide range of AI applications. However, the memory demands posed by the self-attention…

Computation and Language · Computer Science 2023-08-30 Hao Liu , Pieter Abbeel

The impact of transformer networks is booming, yet, they come with significant computational complexity. It is therefore essential to understand how to optimally map and execute these networks on modern neural processor hardware. So far,…

Hardware Architecture · Computer Science 2024-06-17 Steven Colleman , Arne Symons , Victor J. B. Jung , Marian Verhelst

Transformers have become the go-to architecture for language and vision tasks, yet their theoretical properties, especially memorization capacity, remain elusive. This paper investigates the memorization abilities of multi-head attention…

Machine Learning · Computer Science 2024-03-05 Sadegh Mahdavi , Renjie Liao , Christos Thrampoulidis

Instructing the model to generate a sequence of intermediate steps, a.k.a., a chain of thought (CoT), is a highly effective method to improve the accuracy of large language models (LLMs) on arithmetics and symbolic reasoning tasks. However,…

Machine Learning · Computer Science 2024-09-24 Zhiyuan Li , Hong Liu , Denny Zhou , Tengyu Ma

The miniaturization of transistors down to 5nm and beyond, plus the increasing complexity of integrated circuits, significantly aggravate short channel effects, and demand analysis and optimization of more design corners and modes.…

Machine Learning · Computer Science 2020-02-14 Mohammad Saeed Abrishami , Massoud Pedram , Shahin Nazarian

Transformers are among the state of the art for many tasks in speech, vision, and natural language processing, among others. Self-attentions, which are crucial contributors to this performance have quadratic computational complexity, which…

Computation and Language · Computer Science 2022-12-21 Roshan Sharma , Bhiksha Raj

The Transformer model is widely successful on many natural language processing tasks. However, the quadratic complexity of self-attention limit its application on long text. In this paper, adopting a fine-to-coarse attention mechanism on…

Computation and Language · Computer Science 2019-11-12 Zihao Ye , Qipeng Guo , Quan Gan , Xipeng Qiu , Zheng Zhang

Transformer-based large language models (LLMs) have demonstrated surprisingly robust performance across a wide range of language-related tasks, including programming language understanding and generation. In this paper, we take the first…

Programming Languages · Computer Science 2025-01-28 Xiyu Zhai , Runlong Zhou , Liao Zhang , Simon Shaolei Du

We study the role of an essential hyper-parameter that governs the training of Transformers for neural machine translation in a low-resource setting: the batch size. Using theoretical insights and experimental evidence, we argue against the…

Computation and Language · Computer Science 2022-03-22 Àlex R. Atrio , Andrei Popescu-Belis

It is a widely known issue that Transformers, when trained on shorter sequences, fail to generalize robustly to longer ones at test time. This raises the question of whether Transformer models are real reasoning engines, despite their…

Machine Learning · Computer Science 2025-04-04 Ruining Li , Gabrijel Boduljak , Jensen , Zhou

Slim attention shrinks the context memory size by 2x for transformer models with MHA (multi-head attention), which can speed up inference by up to 2x for large context windows. Slim attention is an exact, mathematically identical…

Machine Learning · Computer Science 2025-06-04 Nils Graef , Andrew Wasielewski

Recent LLMs have hundreds of billions of parameters consuming vast resources. Furthermore, the so called "AI scaling law" for transformers suggests that the number of parameters must scale linearly with the size of the data. In response, we…

Computation and Language · Computer Science 2026-01-05 B. N. Kausik

Despite being the current de-facto models in most NLP tasks, transformers are often limited to short sequences due to their quadratic attention complexity on the number of tokens. Several attempts to address this issue were studied, either…

Computation and Language · Computer Science 2023-07-19 Amine Abdaoui , Sourav Dutta

Transformers have been successfully used in various fields and are becoming the standard tools in computer vision. However, self-attention, a core component of transformers, has a quadratic complexity problem, which limits the use of…

Computer Vision and Pattern Recognition · Computer Science 2022-06-02 Jiuk Hong , Chaehyeon Lee , Soyoun Bang , Heechul Jung