Semantic Structure in Deep Learning
Open accessThe DOI matches its Crossref record; 37 of the 37 references it cites were found in a registry or an index.
Abstract
Deep learning has recently come to dominate computational linguistics, leading to claims of human-level performance in a range of language processing tasks. Like much previous computational work, deep learning–based linguistic representations adhere to the distributional meaning-in-use hypothesis, deriving semantic representations from word co-occurrence statistics. However, current deep learning methods entail fundamentally new models of lexical and compositional meaning that are ripe for theoretical analysis. Whereas traditional distributional semantics models take a bottom-up approach in which sentence meaning is characterized by explicit composition functions applied to word meanings, new approaches take a top-down approach in which sentence representations are treated as primary and representations of words and syntax are viewed as emergent. This article summarizes our current understanding of how well such representations capture lexical semantics, world knowledge, and composition. The goal is to foster increased collaboration on testing the implications of such representations as general-purpose models of semantics.
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
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
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
Language Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372
Deep 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
Enriching Word Vectors with Subword Information
Bojanowski, P., Grave, E., Joulin, A., & Mikolov, T. (2017) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00051 · Cited by 9,645
Landauer, T. K., & Dumais, S. T. (1997) · Psychological Review · doi:10.1037/0033-295x.104.2.211 · Cited by 6,282
SQuAD: 100,000+ Questions for Machine Comprehension of Text
Rajpurkar, P., Zhang, J., Lopyrev, K., & Liang, P. (2016) · Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d16-1264 · Cited by 6,208
Harris, Z. S. (1954) · WORD · doi:10.1080/00437956.1954.11659520 · Cited by 4,807
GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding
Wang, A., Singh, A., Michael, J., Hill, F., Levy, O., & Bowman, S. (2018) · Proceedings of the 2018 EMNLP Workshop BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP · doi:10.18653/v1/w18-5446 · Cited by 4,222
Connectionism and cognitive architecture: A critical analysis
Fodor, J. A., & Pylyshyn, Z. W. (1988) · Cognition · doi:10.1016/0010-0277(88)90031-5 · Cited by 4,022
Research works printed but not linked: 103
- Adi Y, Kermany E, Belinkov Y, Lavi O, Goldberg Y. 2017.Fine-grained analysis of sentence embeddings using auxiliary prediction tasks. Paper presented at the International Conference on Learning Representations (ICLR 2017), Toulon, Fr., Apr. 24–26
- Adiwardana D, Luong MT, So DR, Hall J, Fiedel N, et al. 2020. Towards a human-like open-domain chatbot. arXiv:2001.09977 [cs.CL] · arXiv
- Baroni M, 2014, Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics, 1, 238
- Baroni M, 2010, Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing (EMNLP), 1183
- 10.1162/tacl_a_00254 · DOI · Paper page
- Bengio Y, 2003, J. Mach. Learn. Res., 3, 1137
- Berant J, 2011, Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, 610
- 10.1146/annurev-linguistics-011619-030303 · DOI · Paper page
- 10.1162/COLI_a_00261 · DOI · Paper page
- Bowman SR, Dahl GE. 2021. What will it take to fix benchmarking in natural language understanding? arXiv:2104.02145 [cs.CL] · DOI · Paper page
- 10.18653/v1/2021.acl-long.146 · DOI · Paper page
- Carlson A, 2010, Proceedings of the Twenty-Fourth AAAI Conference on Artificial Intelligence, 1306
- 10.18653/v1/2020.conll-1.17 · DOI
- 10.18653/v1/2021.acl-long.258 · DOI · Paper page
- 10.18653/v1/P18-1198 · DOI · Paper page
- Cooper R, 1996, Using the framework
- 10.1007/11736790_9 · DOI · Paper page
- Dasgupta I, Guo D, Stuhlmüller A, Gershman SJ, Goodman ND. 2018. Evaluating compositionality in sentence embeddings. arXiv:1802.04302 [cs.CL] · arXiv
- De Marneffe MC, 2019, Proceedings of Sinn und Bedeutung 23, 107
- Dhillon PS, 2011, NIPS'11: Proceedings of the 24th International Conference on Neural Information Processing Systems, 199
- Dumais ST, 1988, Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 281, 10.1145/57167.57214 · DOI · Paper page
- 10.1207/s15516709cog1402_1 · DOI · Paper page
- 10.18653/v1/2020.acl-main.663 · DOI
- Erk K., 2007, Proceedings of the 45th Annual Meeting of the Association of Computational Linguistics, 216
- 10.1002/lnco.362 · DOI · Paper page
- Ettinger A, 2018, Proceedings of the 27th International Conference on Computational Linguistics, 1790
- 10.18653/v1/W16-2524 · DOI · Paper page
- 10.1145/1409360.1409378 · DOI · Paper page
- 10.1007/978-90-481-8847-5_10 · DOI · Paper page
- Firth JR., 1957, Studies in Linguistic Analysis, 1
- Forbes M, Holtzman A, Choi Y. 2019. Do neural language representations learn physical commonsense? arXiv:1908.02899 [cs.CL] · arXiv
- Frege G., 1884, Die Grundlagen der Arithmetik: Eine logisch mathematische Untersuchung über den Begriff der Zahl
- 10.18653/v1/N19-1004 · DOI · Paper page
- 10.18653/v1/2021.acl-long.212 · DOI
- 10.18653/v1/2020.acl-main.365 · DOI
- 10.18653/v1/2020.acl-main.177 · DOI
- Grefenstette E, Sadrzadeh M. 2011. Experimenting with transitive verbs in a DisCoCat. arXiv:1107.3119 [cs.CL] · arXiv
- 10.1098/rstb.2007.2089 · DOI · Paper page
- 10.18653/v1/E17-1006 · DOI
- Hewitt J, 2019, Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, 1, 4129
- 10.1162/tacl_a_00250 · DOI
- Hupkes D, Singh A, Korrel K, Kruszewski G, Bruni E. 2018a. Learning compositionally through attentive guidance. arXiv:1805.09657 [cs.CL] · arXiv
- 10.1613/jair.1.11196 · DOI · Paper page
- Hwang JD, Bhagavatula C, Bras RL, Da J, Sakaguchi K, et al. 2020. COMET-ATOMIC 2020: on symbolic and neural commonsense knowledge graphs. arXiv:2010.05953 [cs.CL] · DOI · Paper page
- 10.18653/v1/2020.blackboxnlp-1.18 · DOI
- 10.18653/v1/W18-5424 · DOI · Paper page
- Jurafsky D, 2009, Speech and Language Processing
- Kann K, 2019, Proceedings of the Society for Computation in Linguistics (SCiL) 2019, 287
- 10.18653/v1/2020.acl-main.698 · DOI · Paper page
- 10.18653/v1/S19-1026 · DOI · Paper page
- Lake B, 2018, Proceedings of the 35th International Conference on Machine Learning (ICML 2018), 2873
- 10.1162/089120103322145324 · DOI
- Lenci A., 2008, Ital. J. Linguist., 20, 1
- 10.1146/annurev-linguistics-030514-125254 · DOI · Paper page
- 10.18653/v1/2020.emnlp-main.557 · DOI · Paper page
- 10.1162/tacl_a_00115 · DOI · Paper page
- 10.1023/B:BTTJ.0000047600.45421.6d · DOI · Paper page
- 10.18653/v1/N19-1112 · DOI · Paper page
- Lovering C, 2021, Predicting inductive biases of pre-trained models
- Lund K, 1995, Proceedings of the 17th Annual Conference of the Cognitive Science Society, 660
- McKenna N, 2020, Proceedings of the Ninth Joint Conference on Lexical and Computational Semantics, 137
- 10.18653/v1/2020.blackboxnlp-1.4 · DOI · Paper page
- Mickus T, 2020, Proceedings of the Society for Computation in Linguistics 2020, 279
- Mikolov T, Chen K, Corrado G, Dean J. 2013. Efficient estimation of word representations in vector space. arXiv:1301.3781 [cs.CL] · arXiv
- 10.1111/j.1551-6709.2010.01106.x · DOI · Paper page
- 10.18653/v1/2020.blackboxnlp-1.7 · DOI
- Nair S, 2020, Proceedings of the Workshop on the Cognitive Aspects of the Lexicon, 129
- 10.18653/v1/W19-2004 · DOI
- 10.18653/v1/W16-2504 · DOI
- Partee B, 1995, Invitation to Cognitive Science, 1, 311
- 10.18653/v1/P16-1204 · DOI
- 10.3115/v1/D14-1162 · DOI · Paper page
- 10.18653/v1/D18-1007 · DOI · Paper page
- 10.18653/v1/S18-2023 · DOI · Paper page
- Potts C., 2020, Medium
- Radford A, Kim JW, Hallacy C, Ramesh A, Goh G, et al. 2021. Learning transferable visual models from natural language supervision. arXiv:2103.00020 [cs.CV] · arXiv
- Radford A, 2020, OpenAI Blog
- Radford A, 2019, Language models are unsupervised multitask learners
- 10.18653/v1/D16-1264 · DOI · Paper page
- Ravichander A, 2020, Proceedings of the Ninth Joint Conference on Lexical and Computational Semantics, 88
- Reif E, 2019, Advances in Neural Information Processing Systems 32 (NeurIPS 2019)
- 10.18653/v1/P18-1079 · DOI · Paper page
- 10.18653/v1/D19-1228 · DOI
- Saxton D, 2019, Analysing mathematical reasoning abilities of neural models
- 10.1162/tacl_a_00277 · DOI
- 10.18653/v1/2021.acl-long.569 · DOI · Paper page
- 10.1109/ICCV.2019.00756 · DOI · Paper page
- Szabó ZG., 2020, Stanford Encyclopedia of Philosophy
- Tenney I, 2019, What do you learn from context? Probing for sentence structure in contextualized word representations
- 10.18653/v1/2020.blackboxnlp-1.25 · DOI
- 10.1613/jair.2934 · DOI · Paper page
- Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, et al. 2017. Attention is all you need. arXiv:1706.03762 [cs.CL] · arXiv
- Veldhoen S, 2016, Proceedings of the Workshop on Cognitive Computation: Integrating Neural and Symbolic Approaches (CoCo 2016)
- 10.18653/v1/2020.emnlp-main.586 · DOI · Paper page
- 10.18653/v1/D19-1221 · DOI · Paper page
- 10.18653/v1/P19-1439 · DOI · Paper page
- Wang A, 2019, Advances in Neural Information Processing Systems 32 (NeurIPS 2019)
- 10.18653/v1/D19-1286 · DOI · Paper page
- 10.18653/v1/2020.emnlp-main.16 · DOI · Paper page
- White AS, 2018, Proceedings of the 48th Annual Meeting of the North East Linguistic Society (NELS 48), 221
- 10.18653/v1/W19-4804 · DOI
- 10.18653/v1/2020.blackboxnlp-1.15 · DOI · Paper page
- 10.18653/v1/2021.findings-acl.201 · DOI
Cited by
Symbols and grounding in large language models
Pavlick, E. (2023) · Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences · doi:10.1098/rsta.2022.0041 · Cited by 95
“Other recent reviews have been published which are more comprehensive [16,20,21], although given the pace of publication on these topics, no article can be exhaustive.”
BackgroundSelf-citationGoldstein, A., Grinstein-Dabush, A., Schain, M., Wang, H., Hong, Z., Aubrey, B., Nastase, S. A., Zada, Z., Ham, E., Feder, A., et al. (2024) · Nature Communications · doi:10.1038/s41467-024-46631-y · Cited by 64
Coelho Mollo, D., & Millière, R. (2026) · Philosophy and the Mind Sciences · doi:10.33735/phimisci.2026.12307 · Cited by 48
Mollo, D. C., & Millière, R. (2023) · arXiv · doi:10.48550/arxiv.2304.01481 · Cited by 47
Unit Testing for Concepts in Neural Networks
Lovering, C., & Pavlick, E. (2022) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00514 · Cited by 12
Self-citation
Metadata
- Authors
- Pavlick, E.
- Venue
- Annual Review of Linguistics
- Publisher
- Annual Reviews
- Published
- 2022-01-14
- Volume
- 8
- Issue
- 1
- Pages
- 447-471
- ISSN
- 2333-9683 (ISSN-L)
- 2333-9691
- OpenAlex
- W3215599159
- Cited by
- 66
- References
- 37
- FWCI
- 0.78
- 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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Semantic Structure in Deep Learning. https://app.cochat.ai/papers/10.1146/annurev-linguistics-031120-122924
- The paper
- Pavlick, E. (2022). Semantic Structure in Deep Learning. Annual Review of Linguistics. https://doi.org/10.1146/annurev-linguistics-031120-122924
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.