Evaluating Transformer's Ability to Learn Mildly Context-Sensitive Languages
Despite the fact that Transformers perform well in NLP tasks, recent studies suggest that self-attention is theoretically limited in learning even some regular and context-free languages. These findings motivated us to think about their implications in modeling natural language, which is hypothesize...
Gespeichert in:
Hauptverfasser: | , |
---|---|
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | Despite the fact that Transformers perform well in NLP tasks, recent studies
suggest that self-attention is theoretically limited in learning even some
regular and context-free languages. These findings motivated us to think about
their implications in modeling natural language, which is hypothesized to be
mildly context-sensitive. We test the Transformer's ability to learn mildly
context-sensitive languages of varying complexities, and find that they
generalize well to unseen in-distribution data, but their ability to
extrapolate to longer strings is worse than that of LSTMs. Our analyses show
that the learned self-attention patterns and representations modeled dependency
relations and demonstrated counting behavior, which may have helped the models
solve the languages. |
---|---|
DOI: | 10.48550/arxiv.2309.00857 |