Progressive Document-level Text Simplification via Large Language Models

2025-01-07

AI Summary

This paper addresses the challenge of simultaneously preserving semantic fidelity and enhancing readability in long-document simplification. We propose the first stage-wise, collaborative, and progressive simplification framework tailored for large language models (LLMs). Methodologically, we introduce a novel three-level simplification paradigm—“document–topic–word”—and systematically design multi-stage prompt coordination, hierarchical task decomposition, consistency-constrained generation, and cross-granularity control mechanisms, explicitly distinguishing simplification from summarization. Compared to conventional word- or sentence-level approaches and single-prompt LLM methods, our framework achieves significant improvements in simplification quality across multiple benchmarks. It maintains global coherence and semantic faithfulness while enabling controlled dimensionality reduction at structural, syntactic, and lexical levels—establishing new state-of-the-art performance.

Technology Category

• Natural Language Processing: (Large) Language Models
• Machine Learning: Large Multimodal Models (LMMs)
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• 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

Research on text simplification has primarily focused on lexical and sentence-level changes. Long document-level simplification (DS) is still relatively unexplored. Large Language Models (LLMs), like ChatGPT, have excelled in many natural language processing tasks. However, their performance on DS tasks is unsatisfactory, as they often treat DS as merely document summarization. For the DS task, the generated long sequences not only must maintain consistency with the original document throughout, but complete moderate simplification operations encompassing discourses, sentences, and word-level simplifications. Human editors employ a hierarchical complexity simplification strategy to simplify documents. This study delves into simulating this strategy through the utilization of a multi-stage collaboration using LLMs. We propose a progressive simplification method (ProgDS) by hierarchically decomposing the task, including the discourse-level, topic-level, and lexical-level simplification. Experimental results demonstrate that ProgDS significantly outperforms existing smaller models or direct prompting with LLMs, advancing the state-of-the-art in the document simplification task.

Problem

• Large Language Models
• Text Simplification
• Readability Enhancement

Innovation

• Progressive Simplification
• Hierarchical Language Processing
• Large Language Model Collaboration