Metraipolis › archive › dc_5b0b2bae11ef
Analyzing Traditional and Neural Approaches to Multilingual Readability Assessment — arXiv:2609.10792
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.10792
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T05:40:35.677Z
- SHA-256 of the archived text
649cc2debda7d801794a81869c165866ca74c559195c97126e2189ee83a015b6- Stable URL of this page
https://metraipolis.com/documento/dc_5b0b2bae11ef
How to cite
Analyzing Traditional and Neural Approaches to Multilingual Readability Assessment — arXiv:2609.10792 — arXiv (en), https://arxiv.org/abs/2609.10792 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_5b0b2bae11ef (Metraipolis archive dc_5b0b2bae11ef, sha256 649cc2debda7d801794a81869c165866ca74c559195c97126e2189ee83a015b6)
BibTeX
@misc{dc5b0b2bae11ef,
title = {Analyzing Traditional and Neural Approaches to Multilingual Readability Assessment — arXiv:2609.10792},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.10792},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_5b0b2bae11ef; sha256 649cc2debda7d801794a81869c165866ca74c559195c97126e2189ee83a015b6},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.10792. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Transformer-based models excel at Automatic Readability Assessment (ARA), yet feature-based models remain in active use because their predictions tie back to linguistic properties. This matters because readability labels are subjective and rater-dependent, so high accuracy on noisy ground truth may reflect surface patterns rather than the linguistic structure that defines difficulty. We test whether transformers internalize the same features as traditional models across Arabic, English, French, Hindi, and Russian using the ReadMe++ dataset. Shapley Additive Explanations (SHAP) identify the features driving traditional classifiers, which we then use as TCAV concept sets to probe multilingual XLM-R and language-specific encoders. Transformers recover surface-length, syntactic, and lexical-diversity signals, and reflect the ordinal CEFR structure of the traditional models. Alignment varies by model family, language, and layer, with language-specific encoders tracking traditional models more clearly than XLM-R. High linear separability does not always imply directional influence, limiting linear probing for count-based readability features. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-11 in cs.CL). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2609.10792 — 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 2026-10-16T05:23:56.977Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-11?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).