Metraipolis › archive › dc_bdf36d712c6e
Are Language Models Script-Aware? — arXiv:2610.08037
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.08037
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T11:23:19.155Z
- SHA-256 of the archived text
fb0f59de43d573c80126c2d613a49758fe6c3efbc25f76238e2570abde568527- Stable URL of this page
https://metraipolis.com/documento/dc_bdf36d712c6e
How to cite
Are Language Models Script-Aware? — arXiv:2610.08037 — arXiv (en), https://arxiv.org/abs/2610.08037 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bdf36d712c6e (Metraipolis archive dc_bdf36d712c6e, sha256 fb0f59de43d573c80126c2d613a49758fe6c3efbc25f76238e2570abde568527)
BibTeX
@misc{dcbdf36d712c6e,
title = {Are Language Models Script-Aware? — arXiv:2610.08037},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.08037},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bdf36d712c6e; sha256 fb0f59de43d573c80126c2d613a49758fe6c3efbc25f76238e2570abde568527},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.08037. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Language models frequently generate outputs in unintended languages or scripts, a phenomenon known as off-target generation. While existing research has focused on language selection, the dimension of script knowledge remains understudied: before any linguistic understanding can occur, users must recognize the graphic symbols in a model's response. We investigate whether Small and Large Language Models (SLMs and LLMs) possess script knowledge by testing them on multi-scriptic languages. Through two complementary experiments, we evaluate whether models (1) adapt their output script to match the input, and (2) follow explicit instructions to generate text in a specified script. The models we tested demonstrate substantial script knowledge: they all achieve a near-perfect Latin script fidelity (more than 98%) and follow script instructions with high frequency. Nevertheless, we notice differences between LLMs and SLMs, with higher scores for LLMs including for non-standard script combinations. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2610.08037 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-07T11:15:39.445Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).