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

2026-02-07

AI Summary

This work addresses the challenge faced by large language models in unsupervised sentence simplification across languages and readability levels, where balancing simplification effectiveness with semantic fidelity remains difficult. The paper introduces, for the first time, a dynamic stepwise simplification mechanism that decomposes the complex task into controllable steps through dynamic path planning, semantic-aware exemplar selection, and chain-of-thought generation augmented with dialogue history. Evaluated on two benchmarks across five languages, the proposed approach significantly improves simplification quality while reducing computational steps by 22–42%. Human evaluations confirm the gains in output quality and further reveal an inherent trade-off between simplification degree and semantic preservation.

Technology Category

• Natural Language Processing: (Large) Language Models
• Planning, Routing, and Scheduling: Planning with Language Models
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Large language models for search
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

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.

Problem

• sentence simplification
• proficiency control
• semantic preservation
• multilingual
• large language models

Innovation

• proficiency-controlled simplification
• step-by-step decomposition
• semantic-aware exemplar selection
• chain-of-thought reasoning
• multilingual unsupervised simplification