Structure-(in)dependent Interpretation of Phrases in Humans and LSTMs
- Cas W. Coopmans (Max Planck Institute for Psycholinguistics)
- Helen de Hoop (Radboud University)
- Karthikeya Kaushik (Max Planck Institute for Psycholinguistics)
- Peter Hagoort (Max Planck Institute for Psycholinguistics)
- Andrea E. Martin (Max Planck Institute for Psycholinguistics)
Abstract
In this study, we compared the performance of a long short-term memory (LSTM) neural network to the behavior of human participants on a language task that requires hierarchically structured knowledge. We show that humans interpret ambiguous noun phrases, such as second blue ball, in line with their hierarchical constituent structure. LSTMs, instead, only do so after unambiguous training, and they do not systematically generalize to novel items. Overall, the results of our simulations indicate that a model can behave hierarchically without relying on hierarchical constituent structure.
Keywords: syntax, hierarchical structure, constituency, LSTM
How to Cite:
Coopmans, C. W., de Hoop, H., Kaushik, K., Hagoort, P. & Martin, A. E., (2021) “Structure-(in)dependent Interpretation of Phrases in Humans and LSTMs”, Society for Computation in Linguistics 4(1), 459-463. doi: https://doi.org/10.7275/mbrg-1h32
Downloads:
Download PDF