Let's Simplify Step by Step: Guiding LLM Towards Multilingual Unsupervised Proficiency-Controlled Sentence Simplification

TL;DR

This paper introduces a step-by-step framework for multilingual sentence simplification using large language models, enhancing control and effectiveness while highlighting the challenge of maintaining semantic fidelity.

Contribution

It proposes a novel decomposition approach with dynamic path planning and semantic-aware exemplar selection to improve multilingual sentence simplification.

Findings

Improved simplification effectiveness across five languages.

Reduced computational steps by 22-42%.

Human evaluation reveals challenges in semantic preservation.

Abstract

Large language models demonstrate limited capability in proficiency-controlled sentence simplification, particularly when simplifying across large readability levels. We propose a framework that decomposes complex simplifications into manageable steps through dynamic path planning, semantic-aware exemplar selection, and chain-of-thought generation with conversation history for coherent reasoning. Evaluation on five languages across two benchmarks shows our approach improves simplification effectiveness while reducing computational steps by 22-42%. Human evaluation confirms the fundamental trade-off between simplification effectiveness and meaning preservation. Notably, even human annotators struggle to agree on semantic preservation judgments, highlighting the inherent complexity of this task. Our work shows that while step-by-step simplification improves control, preserving semantic fidelity during extensive simplification remains an open challenge.