Assessing the Ability of Self-Attention Networks to Learn Word Order
Open accessThe DOI matches its Crossref record; 15 of the 15 references it cites were found in a registry or an index.
Abstract
Self-attention networks (SAN) have attracted a lot of interests due to their high parallelization and strong performance on a variety of NLP tasks, e.g. machine translation.Due to the lack of recurrence structure such as recurrent neural networks (RNN), SAN is ascribed to be weak at learning positional information of words for sequence modeling.However, neither this speculation has been empirically confirmed, nor explanations for their strong performances on machine translation tasks when "lacking positional information" have been explored.To this end, we propose a novel word reordering detection task to quantify how well the word order information learned by SAN and RNN.Specifically, we randomly move one word to another position, and examine whether a trained model can detect both the original and inserted positions.Experimental results reveal that: 1) SAN trained on word reordering detection indeed has difficulty learning the positional information even with the position embedding; and 2) SAN trained on machine translation learns better positional information than its RNN counterpart, in which position embedding plays a critical role.Although recurrence structure make the model more universally-effective on learning word order, learning objectives matter more in the downstream tasks such as machine translation.
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
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
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 · cited 5×
“, 2018), and language representations (Devlin et al., 2019).”
InfluentialDeep Contextualized Word Representations
Peters, M., Neumann, M., Iyyer, M., Gardner, M., Clark, C., Lee, K., & Zettlemoyer, L. (2018) · Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers) · doi:10.18653/v1/n18-1202 · Cited by 11,983
Improved Semantic Representations From Tree-Structured Long Short-Term Memory Networks
Tai, K. S., Socher, R., & Manning, C. D. (2015) · Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 1: Long Papers) · doi:10.3115/v1/p15-1150 · Cited by 3,081
Convolutional Sequence to Sequence Learning
Gehring, J., Auli, M., Grangier, D., Yarats, D., & Dauphin, Y. N. (2017) · arXiv · doi:10.48550/arxiv.1705.03122 · Cited by 2,850 · cited 4×
“To imitate the order of the sequence, Vaswani et al. (2017) deployed position encodings (Gehring et al., 2017) into SAN.”
InfluentialSelf-Attention with Relative Position Representations
Shaw, P., Uszkoreit, J., & Vaswani, A. (2018) · Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 2 (Short Papers) · doi:10.18653/v1/n18-2074 · Cited by 2,505 · cited 3×
“, 2017) is generally deployed to capture sequential information for SAN (Vaswani et al., 2017; Shaw et al., 2018).”
A Structured Self-attentive Sentence Embedding
Lin, Z., Feng, M., Santos, C. N. D., Yu, M., Xiang, B., Zhou, B., & Bengio, Y. (2017) · arXiv · doi:10.48550/arxiv.1703.03130 · Cited by 1,743 · cited 6×
“Self-attention networks (SAN, Parikh et al., 2016; Lin et al., 2017) have shown promising empirical results in a variety of natural language processing (NLP) tasks, such as machine translation (Vaswani et al., 2017), semantic role labelling (Strubell et al., 2018), and language representations…”
InfluentialDiSAN: Directional Self-Attention Network for RNN/CNN-Free Language Understanding
Shen, T., Zhou, T., Long, G., Jiang, J., Pan, S., & Zhang, C. (2018) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v32i1.11941 · Cited by 779
Rationalizing Neural Predictions
Lei, T., Barzilay, R., & Jaakkola, T. (2016) · Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d16-1011 · Cited by 777 · cited 1×
“Recent studies also pointed out that abandoning irrelevant features in source sentence benefits to some downstream NLP tasks (Lei et al., 2016; Yu et al., 2017; Shen et al., 2018b).”
Cited by
Position Information in Transformers: An Overview
Dufter, P., Schmitt, M., & Schütze, H. (2022) · Computational Linguistics · doi:10.1162/coli_a_00445 · Cited by 175
“Yang et al. (2019) evaluate the ability of recovering the original word positions after shuffling some input words.”
BackgroundTheoretical Limitations of Self-Attention in Neural Sequence Models
Hahn, M. (2020) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00306 · Cited by 165
“Yang et al. (2019) experimentally investigated the power of self-attention to extract word order information, finding differences between recurrent and self-attention models; however, these were modulated by the training objective.”
BackgroundThe Impact of Positional Encoding on Length Generalization in Transformers
Kazemnejad, A., Padhi, I., Ramamurthy, K. N., Das, P., & Reddy, S. (2023) · arXiv · doi:10.48550/arxiv.2305.19466 · Cited by 54
On the Ability and Limitations of Transformers to Recognize Formal Languages
Bhattamishra, S., Ahuja, K., & Goyal, N. (2020) · Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2020.emnlp-main.576 · Cited by 48
Rethinking the Value of Transformer Components
Wang, W., & Tu, Z. (2020) · Proceedings of the 28th International Conference on Computational Linguistics · doi:10.18653/v1/2020.coling-main.529 · Cited by 39
Self-citation
Metadata
- Authors
- Yang, B., Wang, L., Wong, D. F., Chao, L. S., & Tu, Z.
- Venue
- Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics
- Published
- 2019
- Pages
- 3635-3644
- OpenAlex
- W2951977278
- Cited by
- 36
- References
- 15
- FWCI
- 2.70
- Type
- conference paper
- 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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Assessing the Ability of Self-Attention Networks to Learn Word Order. https://app.cochat.ai/papers/10.18653/v1/p19-1354
- The paper
- Yang, B., Wang, L., Wong, D. F., Chao, L. S., & Tu, Z. (2019). Assessing the Ability of Self-Attention Networks to Learn Word Order. Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. https://doi.org/10.18653/v1/p19-1354
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.