Gender-Neutral Rewriting in Italian: Models, Approaches, and Trade-offs
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911158207774720 |
|---|---|
| author | Piergentili, Andrea Savoldi, Beatrice Negri, Matteo Bentivogli, Luisa |
| author_facet | Piergentili, Andrea Savoldi, Beatrice Negri, Matteo Bentivogli, Luisa |
| contents | Gender-neutral rewriting (GNR) aims to reformulate text to eliminate unnecessary gender specifications while preserving meaning, a particularly challenging task in grammatical-gender languages like Italian. In this work, we conduct the first systematic evaluation of state-of-the-art large language models (LLMs) for Italian GNR, introducing a two-dimensional framework that measures both neutrality and semantic fidelity to the input. We compare few-shot prompting across multiple LLMs, fine-tune selected models, and apply targeted cleaning to boost task relevance. Our findings show that open-weight LLMs outperform the only existing model dedicated to GNR in Italian, whereas our fine-tuned models match or exceed the best open-weight LLM's performance at a fraction of its size. Finally, we discuss the trade-off between optimizing the training data for neutrality and meaning preservation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_13480 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Gender-Neutral Rewriting in Italian: Models, Approaches, and Trade-offs Piergentili, Andrea Savoldi, Beatrice Negri, Matteo Bentivogli, Luisa Computation and Language Gender-neutral rewriting (GNR) aims to reformulate text to eliminate unnecessary gender specifications while preserving meaning, a particularly challenging task in grammatical-gender languages like Italian. In this work, we conduct the first systematic evaluation of state-of-the-art large language models (LLMs) for Italian GNR, introducing a two-dimensional framework that measures both neutrality and semantic fidelity to the input. We compare few-shot prompting across multiple LLMs, fine-tune selected models, and apply targeted cleaning to boost task relevance. Our findings show that open-weight LLMs outperform the only existing model dedicated to GNR in Italian, whereas our fine-tuned models match or exceed the best open-weight LLM's performance at a fraction of its size. Finally, we discuss the trade-off between optimizing the training data for neutrality and meaning preservation. |
| title | Gender-Neutral Rewriting in Italian: Models, Approaches, and Trade-offs |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2509.13480 |