English
Related papers

Related papers: Finetuning Pretrained Transformers into RNNs

200 papers

Transformers are state-of-the-art models for a variety of sequence modeling tasks. At their core is an attention function which models pairwise interactions between the inputs at every timestep. While attention is powerful, it does not…

Computation and Language · Computer Science 2021-03-23 Hao Peng , Nikolaos Pappas , Dani Yogatama , Roy Schwartz , Noah A. Smith , Lingpeng Kong

The advent of Transformers marked a significant breakthrough in sequence modelling, providing a highly performant architecture capable of leveraging GPU parallelism. However, Transformers are computationally expensive at inference time,…

Machine Learning · Computer Science 2024-05-29 Leo Feng , Frederick Tung , Hossein Hajimirsadeghi , Mohamed Osama Ahmed , Yoshua Bengio , Greg Mori

We propose an efficient transfer learning method for adapting ImageNet pre-trained Convolutional Neural Network (CNN) to fine-grained image classification task. Conventional transfer learning methods typically face the trade-off between…

Computer Vision and Pattern Recognition · Computer Science 2019-06-13 Xiangxi Mo , Ruizhe Cheng , Tianyi Fang

The original softmax-based attention mechanism (regular attention) in the extremely successful Transformer architecture computes attention between $N$ tokens, each embedded in a $D$-dimensional head, with a time complexity of $O(N^2D)$.…

Machine Learning · Computer Science 2025-10-28 Armin Gerami , Ramani Duraiswami

Recurrent neural networks (RNNs) have represented for years the state of the art in neural machine translation. Recently, new architectures have been proposed, which can leverage parallel computation on GPUs better than classical RNNs.…

Computation and Language · Computer Science 2018-05-14 Mattia Antonino Di Gangi , Marcello Federico

Linear transformers have emerged as a subquadratic-time alternative to softmax attention and have garnered significant interest due to their fixed-size recurrent state that lowers inference cost. However, their original formulation suffers…

Computation and Language · Computer Science 2024-05-13 Jean Mercat , Igor Vasiljevic , Sedrick Keh , Kushal Arora , Achal Dave , Adrien Gaidon , Thomas Kollar

We consider small-data, large-scale decision problems in which a firm must make many operational decisions simultaneously (e.g., across a large product portfolio) while observing only a few, potentially noisy, data points per instance.…

Machine Learning · Computer Science 2026-02-04 Zishi Zhang , Jinhui Han , Ming Hu , Yijie Peng

The pretrain-finetune paradigm usually improves downstream performance over training a model from scratch on the same task, becoming commonplace across many areas of machine learning. While pretraining is empirically observed to be…

Computer Vision and Pattern Recognition · Computer Science 2023-07-13 Gabriele Merlin , Vedant Nanda , Ruchit Rawal , Mariya Toneva

Transformer-based pre-trained models have gained much advance in recent years, becoming one of the most important backbones in natural language processing. Recent work shows that the attention mechanism inside Transformer may not be…

Computation and Language · Computer Science 2022-10-27 Yile Wang , Linyi Yang , Zhiyang Teng , Ming Zhou , Yue Zhang

Since its introduction, softmax attention has become the backbone of modern transformer architectures due to its expressiveness and scalability across a wide range of tasks. However, the main drawback of softmax attention is the quadratic…

Machine Learning · Computer Science 2026-02-20 Gabriel Mongaras , Eric C. Larson

Autoregressive Transformers are strong language models but incur O(T) complexity during per-token generation due to the self-attention mechanism. Recent work proposes kernel-based methods to approximate causal self-attention by replacing it…

Machine Learning · Computer Science 2022-10-11 Huanru Henry Mao

Transformer-based language models utilize the attention mechanism for substantial performance improvements in almost all natural language processing (NLP) tasks. Similar attention structures are also extensively studied in several other…

Computation and Language · Computer Science 2023-05-17 Nurullah Sevim , Ege Ozan Özyedek , Furkan Şahinuç , Aykut Koç

Transformers have become foundational architectures for both natural language and computer vision tasks. However, the high computational cost makes it quite challenging to deploy on resource-constraint devices. This paper investigates the…

Computer Vision and Pattern Recognition · Computer Science 2024-06-18 Jialong Guo , Xinghao Chen , Yehui Tang , Yunhe Wang

Fine-tuning a pretrained transformer for a downstream task has become a standard method in NLP in the last few years. While the results from these models are impressive, applying them can be extremely computationally expensive, as is…

Computation and Language · Computer Science 2020-08-18 Davis Yoshida , Allyson Ettinger , Kevin Gimpel

The intrinsic difficulty in adapting deep learning models to non-stationary environments limits the applicability of neural networks to real-world tasks. This issue is critical in practical supervised learning settings, such as the ones in…

Machine Learning · Computer Science 2023-06-09 Simone Marullo , Matteo Tiezzi , Marco Gori , Stefano Melacci , Tinne Tuytelaars

Recent studies of the computational power of recurrent neural networks (RNNs) reveal a hierarchy of RNN architectures, given real-time and finite-precision assumptions. Here we study auto-regressive Transformers with linearised attention,…

Machine Learning · Computer Science 2023-10-26 Kazuki Irie , Róbert Csordás , Jürgen Schmidhuber

Graph transformers have gained popularity in various graph-based tasks by addressing challenges faced by traditional Graph Neural Networks. However, the quadratic complexity of self-attention operations and the extensive layering in graph…

Machine Learning · Computer Science 2023-09-20 Reza Shirkavand , Heng Huang

In recent years, pretrained models revolutionized the paradigm of natural language understanding (NLU), where we append a randomly initialized classification head after the pretrained backbone, e.g. BERT, and finetune the whole model. As…

Computation and Language · Computer Science 2023-03-29 Zhuoyi Yang , Ming Ding , Yanhui Guo , Qingsong Lv , Jie Tang

Recurrent Neural Networks were, until recently, one of the best ways to capture the timely dependencies in sequences. However, with the introduction of the Transformer, it has been proven that an architecture with only attention-mechanisms…

Machine Learning · Computer Science 2021-08-19 Radostin Cholakov , Todor Kolev

State-of-the-art results on neural machine translation often use attentional sequence-to-sequence models with some form of convolution or recursion. Vaswani et al. (2017) propose a new architecture that avoids recurrence and convolution…

Artificial Intelligence · Computer Science 2017-11-08 Karim Ahmed , Nitish Shirish Keskar , Richard Socher