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

TL;DR

This study evaluates prompting strategies for cross-lingual text simplification between English and French using large language models, focusing on translation, simplification, and their combinations across various datasets.

Contribution

It systematically compares different prompting approaches for CLTS, revealing trade-offs between translation fidelity and simplicity in LLM outputs.

Findings

Direct prompts yield highest BLEU scores for meaning preservation.

Translate-then-Simplify approaches produce the simplest outputs.

Evaluation includes automatic metrics, linguistic analysis, and human judgment.

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 Composition approaches that either translate-then-simplify or simplify-then-translate within a single prompt, and two decomposition 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.