Attention in Natural Language Processing
Open accessThe DOI matches its Crossref record; 176 of the 176 references it cites were found in a registry or an index.
Abstract
Attention is an increasingly popular mechanism used in a wide range of neural architectures. The mechanism itself has been realized in a variety of formats. However, because of the fast-paced advances in this domain, a systematic overview of attention is still missing. In this article, we define a unified model for attention architectures in natural language processing, with a focus on those designed to work with vector representations of the textual data. We propose a taxonomy of attention models according to four dimensions: the representation of the input, the compatibility function, the distribution function, and the multiplicity of the input and/or output. We present the examples of how prior information can be exploited in attention models and discuss ongoing research efforts and open challenges in the area, providing the first extensive categorization of the vast body of literature in this exciting domain.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017) · arXiv · doi:10.48550/arxiv.1706.03762 · Cited by 120,404
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
Exploiting Generative AI to Scale up Intelligent Tutoring Systems
Jakubův, J., Chvalovský, K., Goertzel, Z., Kaliszyk, C., Olšák, M., Piotrowski, B., Schulz, S., Suda, M., & Urban, J. (2023) · DROPS (Schloss Dagstuhl – Leibniz Center for Informatics) · doi:10.4230/lipics.itp.2023.19 · Cited by 76,658
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2018) · arXiv · doi:10.48550/arxiv.1810.04805 · Cited by 73,173
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1423 · Cited by 54,325
Neural Machine Translation by Jointly Learning to Align and Translate
Bahdanau, D., Cho, K., & Bengio, Y. (2014) · arXiv · doi:10.48550/arxiv.1409.0473 · Cited by 16,213
An Introduction to Statistical Learning
James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013) · Springer Texts in Statistics · doi:10.1007/978-1-4614-7138-7 · Cited by 12,106
Rudin, C. (2019) · Nature Machine Intelligence · doi:10.1038/s42256-019-0048-x · Cited by 10,561 · cited 1×
“Their conclusion is that attention may indeed provide an explanation of a model, if by explanation, we mean a plausible, but not necessarily faithful, reconstruction of the decision-making process, as suggested by Rudin [151] and Riedl [152].”
ResultBidirectional recurrent neural networks
Schuster, M., & Paliwal, K. K. (1997) · IEEE Transactions on Signal Processing · doi:10.1109/78.650093 · Cited by 10,172 · cited 1×
“The encoder is a bidirectional recurrent neural network (BiRNN) [133] that computes an annotation term hi for every Fig.”
BackgroundLearning long-term dependencies with gradient descent is difficult
Bengio, Y., Simard, P., & Frasconi, P. (1994) · IEEE Transactions on Neural Networks · doi:10.1109/72.279181 · Cited by 8,542 · cited 1×
“This is especially interesting since it could overcome a well-known shortcoming of RNNs: their limited ability of modeling long-range dependences [140].”
Background
Cited by
A review on the attention mechanism of deep learning
Niu, Z., Zhong, G., & Yu, H. (2021) · Neurocomputing · doi:10.1016/j.neucom.2021.03.091 · Cited by 3,379
An Attentive Survey of Attention Models
Chaudhari, S., Mithal, V., Polatkan, G., & Ramanath, R. (2021) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3465055 · Cited by 689
A Survey of Visual Transformers
Liu, Y., Zhang, Y., Wang, Y., Hou, F., Yuan, J., Tian, J., Zhang, Y., Shi, Z., Fan, J., & He, Z. (2024) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2022.3227717 · Cited by 549
Attention mechanism in neural networks: where it comes and where it goes
Soydaner, D. (2022) · Neural Computing and Applications · doi:10.1007/s00521-022-07366-3 · Cited by 414
A review of deep learning techniques for speech processing
Mehrish, A., Majumder, N., Bharadwaj, R., Mihalcea, R., & Poria, S. (2023) · Information Fusion · doi:10.1016/j.inffus.2023.101869 · Cited by 389
Classification of Remote Sensing Images Using EfficientNet-B3 CNN Model With Attention
Alhichri, H., Alswayed, A. S., Bazi, Y., Ammour, N., & Alajlan, N. A. (2021) · IEEE Access · doi:10.1109/access.2021.3051085 · Cited by 357
“The attention mechanism (the idea of focusing on spe-cific parts of the input) has been applied in deep learning for speech recognition [30], Natural Language processing [31], multimodal reasoning and matching [32], object detection [33], and image recognition [34]–[36].”
BackgroundAttention, please! A survey of neural attention models in deep learning
de Santana Correia, A., & Colombini, E. L. (2022) · Artificial Intelligence Review · doi:10.1007/s10462-022-10148-x · Cited by 298
CNN-Based Transformer Model for Fault Detection in Power System Networks
Thomas, J. B., Chaudhari, S. G., K. V., S., & Verma, N. K. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3238059 · Cited by 275
Change Detection Methods for Remote Sensing in the Last Decade: A Comprehensive Review
Cheng, G., Huang, Y., Li, X., Lyu, S., Xu, Z., Zhao, H., Zhao, Q., & Xiang, S. (2024) · Remote Sensing · doi:10.3390/rs16132355 · Cited by 263
Lv, H., Chen, J., Pan, T., Zhang, T., Feng, Y., & Liu, S. (2022) · Measurement · doi:10.1016/j.measurement.2022.111594 · Cited by 261
Learning to Deceive with Attention-Based Explanations
Pruthi, D., Gupta, M., Dhingra, B., Neubig, G., & Lipton, Z. C. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.432 · Cited by 203
Multimodal data integration for oncology in the era of deep neural networks: a review
Waqas, A., Tripathi, A., Ramachandran, R. P., Stewart, P. A., & Rasool, G. (2024) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2024.1408843 · Cited by 165
“Transformers can handle input sequences of different modalities in a unified way, using the same self-attention mechanism, which processes the inputs as a fully connected graph (Xu et al., 2023).”
MethodA Practical Survey on Faster and Lighter Transformers
Fournier, Q., Caron, G. M., & Aloise, D. (2023) · ACM Computing Surveys · doi:10.1145/3586074 · Cited by 145
“idden representations h i for i= 1;:::;nwhere the weights are computed with a feed-forward neural network. For a comprehensive survey of the attention mechanism, we refer the reader to Galassi et al. [7] and Weng [8]. Figure 3 illustrates the inter-attention framework. Moreover, recurrent neural networks do not scale…”
MethodIs GitHub’s Copilot as bad as humans at introducing vulnerabilities in code?
Asare, O., Nagappan, M., & Asokan, N. (2023) · Empirical Software Engineering · doi:10.1007/s10664-023-10380-1 · Cited by 128
A channel-spatial-temporal attention-based network for vibration-based damage detection
Liao, S., Liu, H., Yang, J., & Ge, Y. (2022) · Information Sciences · doi:10.1016/j.ins.2022.05.042 · Cited by 31
Rami Naidji, M., & Elberrichi, Z. (2024) · International Journal of Computing and Digital Systems · doi:10.12785/ijcds/1501102 · Cited by 4
Metadata
- Authors
- Galassi, A., Lippi, M., & Torroni, P.
- Venue
- IEEE Transactions on Neural Networks and Learning Systems
- Published
- 2021-10
- Volume
- 32
- Issue
- 10
- Pages
- 4291-4308
- ISSN
- 2162-237X (ISSN-L)
- 2162-2388
- OpenAlex
- W3031696893
- Cited by
- 635
- References
- 176
- FWCI
- 31.35
- Type
- journal article
- Language
- en
- Open access
- hybrid
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.