04:00
2026-09-11
arxiv.org
natural-language-processing
Analyzing Traditional and Neural Approaches to Multilingual Readability Assessment
A new arXiv paper (2609.10792v1) tests whether transformer-based models internalize the same linguistic features as traditional feature-based classifiers in Automatic Readability Assessment across Ara…