# Can LLMs Control Readability? A Multi-Dimensional Evaluation Framework for CEFR-Controlled Arabic Generation

arXiv: https://arxiv.org/abs/2606.21981 · HTML: https://mlsanity.com/paper/2606.21981/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 19 references that are in the map.
- First posted on arXiv 20 Jun 2026.

## Builds on (most cited first, 8 of 8)

- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [Text Readability Assessment for Second Language Learners](https://mlsanity.com/paper/1906.07580/) (2019)
- [Generating Summaries with Controllable Readability Levels](https://mlsanity.com/paper/2310.10623/) (2023)
- [ReadMe++: Benchmarking Multilingual Language Models for Multi-Domain Readability Assessment](https://mlsanity.com/paper/2305.14463/) (2023)
- [Measuring and Modifying the Readability of English Texts with GPT-4](https://mlsanity.com/paper/2410.14028/) (2024)
- [A Large and Balanced Corpus for Fine-grained Arabic Readability Assessment](https://mlsanity.com/paper/2502.13520/) (2025)
- [Automatic Difficulty Classification of Arabic Sentences](https://mlsanity.com/paper/2103.04386/) (2021)
- [Enhancing Arabic Automated Essay Scoring with Synthetic Data and Error Injection](https://mlsanity.com/paper/2503.17739/) (2025)

## Abstract (arXiv)

While Large Language Models (LLMs) can generate fluent Arabic text, their ability to reliably control readability levels remains unclear. We propose a multi-dimensional evaluation framework for Common European Framework of Reference for Language (CEFR)-controlled Arabic text generation, assessing whether instruction-following LLMs can serve as reliable generators for adaptive language learning. Our framework integrates controlled prompting, automatic readability prediction using a validated Taha-19 model, lexical constraint validation, and syntactic complexity profiling. Results show that structured prompting substantially improves CEFR alignment. In particular, CEFR-guided prompting with lexical constraints achieves the highest conformity to reference linguistic profiles (0.91 cosine similarity) and near-perfect agreement with predicted readability levels (0.99), while unconstrained prompting exhibits weak control. These findings establish an empirical foundation for integrating readability-aware Arabic text generation into adaptive educational systems.