A Ship of Theseus: Curious Cases of Paraphrasing in LLM-Generated Texts

TL;DR

This paper explores whether texts retain their original authorship after multiple paraphrasing iterations by LLMs, revealing that style deviation affects authorship attribution and challenging traditional notions of authorship.

Contribution

It introduces a computational method to analyze authorship in paraphrased texts and demonstrates how style deviation impacts authorship attribution in LLM-generated content.

Findings

Performance of text classification models decreases with each paraphrasing iteration.

Extent of style deviation correlates with authorship attribution challenges.

Reconsideration of authorship notions in the context of AI paraphrasing.

Abstract

In the realm of text manipulation and linguistic transformation, the question of authorship has been a subject of fascination and philosophical inquiry. Much like the Ship of Theseus paradox, which ponders whether a ship remains the same when each of its original planks is replaced, our research delves into an intriguing question: Does a text retain its original authorship when it undergoes numerous paraphrasing iterations? Specifically, since Large Language Models (LLMs) have demonstrated remarkable proficiency in both the generation of original content and the modification of human-authored texts, a pivotal question emerges concerning the determination of authorship in instances where LLMs or similar paraphrasing tools are employed to rephrase the text--i.e., whether authorship should be attributed to the original human author or the AI-powered tool. Therefore, we embark on a philosophical voyage through the seas of language and authorship to unravel this intricate puzzle. Using a computational approach, we discover that the diminishing performance in text classification models, with each successive paraphrasing iteration, is closely associated with the extent of deviation from the original author's style, thus provoking a reconsideration of the current notion of authorship.