Search papers, labs, and topics across Lattice.
This paper introduces an iterative prompt refinement pipeline leveraging GPT-4o to generate dyslexia-friendly text summaries, targeting a Flesch Reading Ease score >= 90. The pipeline refines prompts iteratively to improve readability while maintaining semantic fidelity, evaluated on a dataset of 2,000 news articles. Results demonstrate that the approach achieves the target readability within a few iterations, establishing a baseline for accessibility-driven NLP summarization.
GPT-4o can iteratively refine text summaries to be dyslexia-friendly with surprising efficiency, achieving high readability scores in just a few attempts.
Dyslexia affects approximately 10% of the global population and presents persistent challenges in reading fluency and text comprehension. While existing assistive technologies address visual presentation, linguistic complexity remains a substantial barrier to equitable access. This paper presents an empirical study on dyslexia-friendly text summarization using an iterative prompt-based refinement pipeline built on GPT-4o. We evaluate the pipeline on approximately 2,000 news article samples, applying a readability target of Flesch Reading Ease>= 90. Results show that the majority of summaries meet the readability threshold within four attempts, with many succeeding on the first try. A composite score combining readability and semantic fidelity shows stable performance across the dataset, ranging from 0.13 to 0.73 with a typical value near 0.55. These findings establish an empirical baseline for accessibility-driven NLP summarization and motivate further human-centered evaluation with dyslexic readers.
