Hierarchical Representation in Neural Language Models: Suppression and Recovery of Expectations
Open accessThe DOI matches its Crossref record; 8 of the 8 references it cites were found in a registry or an index.
Abstract
Deep learning sequence models have led to a marked increase in performance for a range of Natural Language Processing tasks, but it remains an open question whether they are able to induce proper hierarchical generalizations for representing natural language from linear input alone.Work using artificial languages as training input has shown that LSTMs are capable of inducing the stack-like data structures required to represent context-free and certain mildly context-sensitive languages (Weiss et al., 2018)-formal language classes which correspond in theory to the hierarchical structures of natural language.Here we present a suite of experiments probing whether neural language models trained on linguistic data induce these stack-like data structures and deploy them while incrementally predicting words.We study two natural language phenomena: center embedding sentences and syntactic island constraints on the filler-gap dependency.In order to properly predict words in these structures, a model must be able to temporarily suppress certain expectations and then recover those expectations later, essentially pushing and popping these expectations on a stack.Our results provide evidence that models can successfully suppress and recover expectations in many cases, but do not fully recover their previous grammatical state.
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
Hochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318
Distributed representations, simple recurrent networks, and grammatical structure
Elman, J. L. (1991) · Machine Learning · doi:10.1007/bf00114844 · Cited by 1,354
A probabilistic earley parser as a psycholinguistic model
Hale, J. (2001) · Second meeting of the North American Chapter of the Association for Computational Linguistics on Language technologies 2001 - NAACL '01 · doi:10.3115/1073336.1073357 · Cited by 1,150
One Billion Word Benchmark for Measuring Progress in Statistical Language Modeling
Chelba, C., Mikolov, T., Schuster, M., Ge, Q., Brants, T., Koehn, P., & Robinson, T. (2013) · arXiv · doi:10.48550/arxiv.1312.3005 · Cited by 1,051
The effect of word predictability on reading time is logarithmic
Smith, N. J., & Levy, R. (2013) · Cognition · doi:10.1016/j.cognition.2013.02.013 · Cited by 725
One billion word benchmark for measuring progress in statistical language modeling
Chelba, C., Mikolov, T., Schuster, M., Ge, Q., Brants, T., Koehn, P., & Robinson, T. (2014) · Interspeech 2014 · doi:10.21437/interspeech.2014-564 · Cited by 627
Recurrent Neural Network Grammars
Dyer, C., Kuncoro, A., Ballesteros, M., & Smith, N. A. (2016) · Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies · doi:10.18653/v1/n16-1024 · Cited by 517 · cited 4×
“Answering this question is important both for technical outcomes—models with explicit hierarchical structure show performance gains, at least when training on relatively small datasets (Choe and Charniak, 2016; Dyer et al., 2016; Kuncoro et al., 2016)—and for the scientific aim of understanding what biases, learn-”
BackgroundMethodInfluentialOn the Practical Computational Power of Finite Precision RNNs for Language Recognition
Weiss, G., Goldberg, Y., & Yahav, E. (2018) · Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers) · doi:10.18653/v1/p18-2117 · Cited by 226
Cited by
Lack of selectivity for syntax relative to word meanings throughout the language network
Fedorenko, E., Blank, I. A., Siegelman, M., & Mineroff, Z. (2020) · Cognition · doi:10.1016/j.cognition.2020.104348 · Cited by 170
A Systematic Assessment of Syntactic Generalization in Neural Language Models
Hu, J., Gauthier, J., Qian, P., Wilcox, E., & Levy, R. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.158 · Cited by 134
Self-citationLocalizing syntactic predictions using recurrent neural network grammars
Brennan, J. R., Dyer, C., Kuncoro, A., & Hale, J. T. (2020) · Neuropsychologia · doi:10.1016/j.neuropsychologia.2020.107479 · Cited by 58
Lampinen, A. (2024) · Computational Linguistics · doi:10.1162/coli_a_00525 · Cited by 32
“…syntactic structures (e.g., Futrell et al. 2019; Hu et al. 2020; Wilcox et al. 2020), and in particular recursive dependencies (Futrell et al. 2018; Wilcox, Levy, and Futrell 2019; Lakretz et al. 2021)—and even whether they can be used to help model human processing of these structures (Hahn et…”
Lampinen, A. K. (2022) · arXiv · doi:10.48550/arxiv.2210.15303 · Cited by 32
Metadata
- Authors
- Wilcox, E., Levy, R., & Futrell, R.
- Venue
- Proceedings of the 2019 ACL Workshop BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP
- Published
- 2019
- Pages
- 181-190
- OpenAlex
- W2948201212
- Cited by
- 21
- References
- 8
- FWCI
- 1.90
- Type
- conference paper
- Language
- en
- Open access
- gold
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: Hierarchical Representation in Neural Language Models: Suppression and Recovery of Expectations. https://app.cochat.ai/papers/10.18653/v1/w19-4819
- The paper
- Wilcox, E., Levy, R., & Futrell, R. (2019). Hierarchical Representation in Neural Language Models: Suppression and Recovery of Expectations. Proceedings of the 2019 ACL Workshop BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP. https://doi.org/10.18653/v1/w19-4819
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.