Metraipolis › archive › dc_fea4f411dc59
Contextual Embedding Evidence for Main--Light Verb Distinctions in Urdu — arXiv:2608.23645
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.23645
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-18T06:17:51.609Z
- SHA-256 of the archived text
a3f04c67870aef7158ec85de9eccb86a7f8b903e1bc225a6d5995aa9baa3ffc4- Stable URL of this page
https://metraipolis.com/documento/dc_fea4f411dc59
How to cite
Contextual Embedding Evidence for Main--Light Verb Distinctions in Urdu — arXiv:2608.23645 — arXiv (en), https://arxiv.org/abs/2608.23645 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fea4f411dc59 (Metraipolis archive dc_fea4f411dc59, sha256 a3f04c67870aef7158ec85de9eccb86a7f8b903e1bc225a6d5995aa9baa3ffc4)
BibTeX
@misc{dcfea4f411dc59,
title = {Contextual Embedding Evidence for Main--Light Verb Distinctions in Urdu — arXiv:2608.23645},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.23645},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fea4f411dc59; sha256 a3f04c67870aef7158ec85de9eccb86a7f8b903e1bc225a6d5995aa9baa3ffc4},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.23645. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Urdu light verbs contribute schematic event-structural meaning while remaining lexically related to corresponding main verbs. This study tests representational predictions derived from Butt's analysis using contextual embeddings from UrduBERT, DunbaaBERT, and multilingual BERT across 1,126 naturally occurring sentences containing seven Urdu verbs. Main and light uses show significant representational separation in all 21 verb--model comparisons. At the same time, same-lemma main and light centroids are consistently closer than mismatched main--light lemma pairs, supporting continued lexical relatedness. In a seven-way prediction task restricted to light uses, verb identity remains recoverable after the target is masked, with UrduBERT achieving 0.866 accuracy and 0.852 macro-F1. UrduBERT also retains 0.782 accuracy under a preceding-form-disjoint evaluation, indicating generalization beyond repeated local verb combinations. These findings provide computational evidence consistent with Butt's account that Urdu light verbs differ systematically from their main uses while retaining lemma-specific and verb-specific representational structure. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-26 in cs.CL). Categorie arXiv: Computation and Language (cs.CL) ; Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2608.23645 — 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 2026-10-18T06:17:29.336Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-26?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).