Â鶹ÊÓƵ

The research explores how to develop a systematic approach for mapping text simplification guidelines and user requirements to structured prompts that enable automatic text simplification using Large Language Models (LLMs). It will analyse the completeness and feasibility of such mappings across various guideline sets. To begin with, simplifications will be explored at different linguistic levels (e.g., syntactic, lexical) to refine prompt development. Subsequently, sequences of prompts will be applied to ensure comprehensive text transformations tailored to user needs. The study will feature multiple use cases to assess the adaptability of the proposed method. Key scenarios include improving accessibility for non-native speakers (e.g., simplifying legal or medical texts) and adapting content for individuals with learning disabilities, involving feedback from carers. Additionally, collaborations with local government bodies may facilitate applications in making public information more accessible. Evaluation methods will include reading comprehension tests and qualitative assessments from stakeholders.