Simplify-This: A Comparative Analysis of Prompt-Based and Fine-Tuned LLMs

TL;DR

This paper compares prompt-based and fine-tuned large language models for text simplification, finding fine-tuning generally produces better structural simplification while prompting achieves higher semantic similarity, supported by human evaluations.

Contribution

It provides a comprehensive comparison of prompt-based and fine-tuned LLMs for text simplification across multiple benchmarks and metrics, including new datasets and models for reproducibility.

Findings

Fine-tuned models excel in structural simplification.

Prompting often yields higher semantic similarity scores.

Human evaluation favors fine-tuned outputs.

Abstract

Large language models (LLMs) enable strong text generation, and in general there is a practical tradeoff between fine-tuning and prompt engineering. We introduce Simplify-This, a comparative study evaluating both paradigms for text simplification with encoder-decoder LLMs across multiple benchmarks, using a range of evaluation metrics. Fine-tuned models consistently deliver stronger structural simplification, whereas prompting often attains higher semantic similarity scores yet tends to copy inputs. A human evaluation favors fine-tuned outputs overall. We release code, a cleaned derivative dataset used in our study, checkpoints of fine-tuned models, and prompt templates to facilitate reproducibility and future work.