LLM-based Text Simplification and its Effect on User Comprehension and Cognitive Load
Topics: Data Mining, Gemini, LLMO / GEO
This Google affiliated paper presents a large-scale randomized study evaluating how a self-refinement approach to LLM-based text simplification affects reader comprehension and cognitive load. Across six domains—including biomedical articles, finance, and law—over 4,500 participants read either original or automatically simplified passages and answered comprehension questions. The results show a statistically significant improvement in accuracy (3.9% absolute increase overall, with gains up to 14.6% in biomedical texts) and a reduction in perceived task difficulty for simplified texts.