Do LSTMs See Gender? Probing the Ability of LSTMs to Learn Abstract Syntactic Rules
Not on the map. Why?
Abstract (arXiv)
LSTMs trained on next-word prediction can accurately perform linguistic tasks that require tracking long-distance syntactic dependencies. Notably, model accuracy approaches human performance on number agreement tasks (Gulordava et al., 2018). However, we do not have a mechanistic understanding of how LSTMs perform such linguistic tasks. Do LSTMs learn abstract grammatical rules, or do they rely on simple heuristics? Here, we test gender agreement in French which requires tracking both hierarchical syntactic structures and the inherent gender of lexical units. Our model is able to reliably predict long-distance gender agreement in two subject-predicate contexts: noun-adjective and noun-passive-verb agreement. The model showed more inaccuracies on plural noun phrases with gender attractors compared to singular cases, suggesting a reliance on clues from gendered articles for agreement. Overall, our study highlights key ways in which LSTMs deviate from human behaviour and questions whether LSTMs genuinely learn abstract syntactic rules and categories. We propose using gender agreement as a useful probe to investigate the underlying mechanisms, internal representations, and linguistic capabilities of LSTM language models.
Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Builds on 6 of its 8 references that are in the map. First posted on arXiv 31 Oct 2022.
Builds on
6 of 8 references.
References · 6
- 2016Assessing the Ability of LSTMs to Learn Syntax-Sensitive Dependenciescited by 350
- 2018Colorless green recurrent networks dream hierarchicallycited by 210
- 2018Under the Hood: Using Diagnostic Classifiers to Investigate and Improve how Language Models Track Agreement Informationcited by 97
- 2019The emergence of number and syntax units in LSTM language modelscited by 92
- 2020Cross-Linguistic Syntactic Evaluation of Word Prediction Modelscited by 54
- 2020Mechanisms for Handling Nested Dependencies in Neural-Network Language Models and Humanscited by 42
Built on by
Citing papers per year: 2023: 5; 2024: 1.
6 papers.
- 2023Neither hype nor gloom do DNNs justicecited by 40
- 2023Investigating semantic subspaces of Transformer sentence embeddings through linear structural probingcited by 12
- 2024Investigating grammatical abstraction in language models using few-shot learning of novel noun gendercited by 2
- 2023Beyond the limitations of any imaginable mechanism: large language models and psycholinguisticscited by 1
- 2023PatternGPT :A Pattern-Driven Framework for Large Language Model Text Generationcited by 1
- 2023Fixing the problems of deep neural networks will require better training data and learning algorithmscited by 0
- 2024Investigating grammatical abstraction in language models using few-shot learning of novel noun gender15 Mar 2024
- 2023Neither hype nor gloom do DNNs justice8 Dec 2023
- 2023Investigating semantic subspaces of Transformer sentence embeddings through linear structural probing18 Oct 2023
- 2023Fixing the problems of deep neural networks will require better training data and learning algorithms26 Sep 2023
- 2023PatternGPT :A Pattern-Driven Framework for Large Language Model Text Generation2 Jul 2023
- 2023Beyond the limitations of any imaginable mechanism: large language models and psycholinguistics28 Feb 2023