Get Started
Home
Topics
Search
Library
Research questionHow can language models improve accessibility-focused text simplification for low-resource languages when cross-lingual transfer is unreliable?Accessibility-oriented simplification requires preserving meaning while making text easier to understand, yet low-resource languages often lack training data and reliable transfer from better-resourced languages. Standard NLP improvements do not consistently produce useful gains for Catalan simplification.
AI
Evaluation & Benchmarks
LLM Pretraining & Post-training
Machine Learning
Natural Language Processing
Reinforcement Learning
Latest papersRecent research connected to this question, newest first.Reinforcement Learning for improving Large Language Models' Catalan text simplification capabilitiesThe evidence concerns reinforcement-learning post-training of IberianLLM-7B-Instruct for Catalan text simplification, using English ASSET and translated Catalan and Spanish variants. Results were assessed on two curated Catalan benchmarks; translated-data post-training did not significantly improve an out-of-domain benchmark, so the findings do not establish broad cross-lingual or domain transfer.research paper · Sep 4, 2026
Related questions
How can models adapt to low-resource languages without damaging source-language and related-task performance?How can multilingual NLP models reliably detect harmful posts and verifiable factual claims in low-resource languages?How can multilingual representation sharing be measured without confusing anisotropy with genuine cross-lingual structure?How can sparse autoencoder features be shared across language models without per-model retraining?