Translate or Simplify First: An Analysis of Cross-lingual Text Simplification in English and French

ACL ARR 2026 January Submission9335 Authors

Keywords: Text simplification, Cross-lingual text simplification, LLM's

Abstract:

Cross-Lingual Text Simplification (CLTS) aims to make content more accessible across languages by simultaneously addressing both linguistic complexity and translation. This study investigates the effectiveness of different prompting strategies for CLTS between English and French using large language models (LLMs). We examine five distinct prompting systems: a direct prompt instructing the LLM to perform both translation and simplification simultaneously, two chain-of-thought (CoT) approaches that either translate-then-simplify or simplify-then-translate within a single prompt, and two pipeline approaches that perform the same operations in separate, consecutive prompts.
These systems are evaluated across a diverse set of five corpora of different genres (Wikipedia and medical texts) using seven state-of-the-art LLMs. Output quality is assessed through a multi-faceted evaluation framework comprising automatic metrics, comprehensive linguistic feature analysis, and human evaluation of simplicity and meaning preservation. Our findings reveal that while direct prompting consistently achieves the highest BLEU scores, indicating meaning fidelity, Translate-then-Simplify approaches demonstrate the highest simplicity, as measured by the linguistic features.

Paper Type: Long

Research Area: Multilinguality and Language Diversity

Research Area Keywords: cross-lingual transfer, linguistic variation

Contribution Types: NLP engineering experiment, Data resources

Languages Studied: English, French

Submission Number: 9335