English
Related papers

Related papers: Local Self-Attention over Long Text for Efficient …

200 papers

Transformer-based LLMs have achieved exceptional performance across a wide range of NLP tasks. However, the standard self-attention mechanism suffers from quadratic time complexity and linearly increased cache size. Sliding window attention…

Computation and Language · Computer Science 2025-01-03 Yixing Xu , Shivank Nag , Dong Li , Lu Tian , Emad Barsoum

Attention mechanism has been used as an ancillary means to help RNN or CNN. However, the Transformer (Vaswani et al., 2017) recently recorded the state-of-the-art performance in machine translation with a dramatic reduction in training time…

Computation and Language · Computer Science 2017-12-07 Jinbae Im , Sungzoon Cho

The recent literature in text classification is biased towards short text sequences (e.g., sentences or paragraphs). In real-world applications, multi-page multi-paragraph documents are common and they cannot be efficiently encoded by…

Computation and Language · Computer Science 2022-10-26 Xiang Dai , Ilias Chalkidis , Sune Darkner , Desmond Elliott

We present the first unified study of the efficiency of self-attention-based Transformer variants spanning text, speech and vision. We identify input length thresholds (tipping points) at which efficient Transformer variants become more…

Computation and Language · Computer Science 2023-06-16 Anuj Diwan , Eunsol Choi , David Harwath

The multi-head self-attention of popular transformer models is widely used within Natural Language Processing (NLP), including for the task of extractive summarization. With the goal of analyzing and pruning the parameter-heavy…

Computation and Language · Computer Science 2020-12-04 Wen Xiao , Patrick Huber , Giuseppe Carenini

Recent works have revealed that Transformers are implicitly learning the syntactic information in its lower layers from data, albeit is highly dependent on the quality and scale of the training data. However, learning syntactic information…

Computation and Language · Computer Science 2022-10-24 Shengyuan Hou , Jushi Kai , Haotian Xue , Bingyu Zhu , Bo Yuan , Longtao Huang , Xinbing Wang , Zhouhan Lin

The self-attention module is a key component of Transformer-based models, wherein each token pays attention to every other token. Recent studies have shown that these heads exhibit syntactic, semantic, or local behaviour. Some studies have…

Computation and Language · Computer Science 2020-08-14 Madhura Pande , Aakriti Budhraja , Preksha Nema , Pratyush Kumar , Mitesh M. Khapra

Despite the progress made in sentence-level NMT, current systems still fall short at achieving fluent, good quality translation for a full document. Recent works in context-aware NMT consider only a few previous sentences as context and may…

Computation and Language · Computer Science 2019-05-27 Sameen Maruf , André F. T. Martins , Gholamreza Haffari

Despite the success of Transformers, handling long contexts remains challenging due to the limited length generalization and quadratic complexity of self-attention. Thus Transformers often require post-training with a larger attention…

Computation and Language · Computer Science 2025-06-13 Xiang Hu , Zhihao Teng , Jun Zhao , Wei Wu , Kewei Tu

Transformer-based models have achieved dominant performance in numerous NLP tasks. Despite their remarkable successes, pre-trained transformers such as BERT suffer from a computationally expensive self-attention mechanism that interacts…

Computation and Language · Computer Science 2024-06-04 Jungmin Yun , Mihyeon Kim , Youngbin Kim

Deep pretrained transformer networks are effective at various ranking tasks, such as question answering and ad-hoc document ranking. However, their computational expenses deem them cost-prohibitive in practice. Our proposed approach, called…

Information Retrieval · Computer Science 2020-05-27 Sean MacAvaney , Franco Maria Nardini , Raffaele Perego , Nicola Tonellotto , Nazli Goharian , Ophir Frieder

Self-attention mechanism has been a key factor in the recent progress of Vision Transformer (ViT), which enables adaptive feature extraction from global contexts. However, existing self-attention methods either adopt sparse global attention…

Computer Vision and Pattern Recognition · Computer Science 2023-04-11 Xuran Pan , Tianzhu Ye , Zhuofan Xia , Shiji Song , Gao Huang

Transformer-based models are unable to process long sequences due to their self-attention operation, which scales quadratically with the sequence length. To address this limitation, we introduce the Longformer with an attention mechanism…

Computation and Language · Computer Science 2020-12-03 Iz Beltagy , Matthew E. Peters , Arman Cohan

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

In pursuit of faster computation, Efficient Transformers demonstrate an impressive variety of approaches -- models attaining sub-quadratic attention complexity can utilize a notion of sparsity or a low-rank approximation of inputs to reduce…

Machine Learning · Computer Science 2022-11-09 Uladzislau Yorsh , Alexander Kovalenko

Learned image compression methods have exhibited superior rate-distortion performance than classical image compression standards. Most existing learned image compression models are based on Convolutional Neural Networks (CNNs). Despite…

Image and Video Processing · Electrical Eng. & Systems 2022-04-12 Renjie Zou , Chunfeng Song , Zhaoxiang Zhang

Vision Transformers achieved outstanding performance in many computer vision tasks. Early Vision Transformers such as ViT and DeiT adopt global self-attention, which is computationally expensive when the number of patches is large. To…

Computer Vision and Pattern Recognition · Computer Science 2022-10-20 Tan Yu , Gangming Zhao , Ping Li , Yizhou Yu

Cross-encoders are effective passage and document re-rankers but less efficient than other neural or classic retrieval models. A few previous studies have applied windowed self-attention to make cross-encoders more efficient. However, these…

Information Retrieval · Computer Science 2024-03-21 Ferdinand Schlatt , Maik Fröbe , Matthias Hagen

Learning to rank is an important task that has been successfully deployed in many real-world information retrieval systems. Most existing methods compute relevance judgments of documents independently, without holistically considering the…

Information Retrieval · Computer Science 2020-05-11 Shuo Sun , Kevin Duh

Document-level context for neural machine translation (NMT) is crucial to improve the translation consistency and cohesion, the translation of ambiguous inputs, as well as several other linguistic phenomena. Many works have been published…

Computation and Language · Computer Science 2023-06-09 Christian Herold , Hermann Ney