Neural Induction of Finite-State Transducers
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908774967541760 |
|---|---|
| author | Ginn, Michael Palmer, Alexis Hulden, Mans |
| author_facet | Ginn, Michael Palmer, Alexis Hulden, Mans |
| contents | Finite-State Transducers (FSTs) are effective models for string-to-string rewriting tasks, often providing the efficiency necessary for high-performance applications, but constructing transducers by hand is difficult. In this work, we propose a novel method for automatically constructing unweighted FSTs following the hidden state geometry learned by a recurrent neural network. We evaluate our methods on real-world datasets for morphological inflection, grapheme-to-phoneme prediction, and historical normalization, showing that the constructed FSTs are highly accurate and robust for many datasets, substantially outperforming classical transducer learning algorithms by up to 87% accuracy on held-out test sets. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_10918 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Neural Induction of Finite-State Transducers Ginn, Michael Palmer, Alexis Hulden, Mans Computation and Language Finite-State Transducers (FSTs) are effective models for string-to-string rewriting tasks, often providing the efficiency necessary for high-performance applications, but constructing transducers by hand is difficult. In this work, we propose a novel method for automatically constructing unweighted FSTs following the hidden state geometry learned by a recurrent neural network. We evaluate our methods on real-world datasets for morphological inflection, grapheme-to-phoneme prediction, and historical normalization, showing that the constructed FSTs are highly accurate and robust for many datasets, substantially outperforming classical transducer learning algorithms by up to 87% accuracy on held-out test sets. |
| title | Neural Induction of Finite-State Transducers |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2601.10918 |