Abstract:The diachronic evolution from Latin to the Romance languages involved a restructuring of the grammatical gender system from a tripartite configuration (masculine, feminine, neuter) to a bipartite one (masculine, feminine). In this work, we introduce an interpretable deep learning framework to investigate this phenomenon at both lexical and contextual levels. First, we show that conventional tokenization strategies are insufficiently robust for this low-resource historical setting, and that our proposed tokenizer improves performance over these baselines. At the lexical level, we evaluate the contribution of morphological features to gender prediction. At the contextual level, we quantify the contributions of different part-of-speech categories to grammatical gender prediction. Together, these analyses characterize the distribution of gender information between the lemma and its sentential context. We make our codebase, datasets, and results publicly available.
| Comments: | Accepted at NLP4DH @ ACL 2026 |
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2605.09156 [cs.CL] |
| (or arXiv:2605.09156v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2605.09156 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Esteban Garces Arias [view email]
[v1]
Sat, 9 May 2026 20:36:49 UTC (6,218 KB)