mlsanityData to 14 Oct 2026

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

arXiv 2311.08374 ↗First posted 14 Nov 2023cs.CLcited by 23

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. Cited by 23 papers in the map (data through 14 Oct 2026). Builds on 38 of its 72 references that are in the map. First posted on arXiv 14 Nov 2023.

Builds on

38 of 72 references.

References · 38

  1. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  2. 2019BERTScore: Evaluating Text Generation with BERTcited by 5,512
  3. 2016XGBoost: A Scalable Tree Boosting Systemcited by 5,166
  4. 2018GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understandingcited by 4,899
  5. 2015Character-level Convolutional Networks for Text Classificationcited by 2,330
  6. 2022BLOOM: A 176B-Parameter Open-Access Multilingual Language Modelcited by 1,984
  7. 2018Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarizationcited by 1,317
  8. 2018Hierarchical Neural Story Generationcited by 1,193
  9. 2019PEGASUS: Pre-training with Extracted Gap-sentences for Abstractive Summarizationcited by 1,102
  10. 2022GLM-130B: An Open Bilingual Pre-trained Modelcited by 843
  11. 2023DetectGPT: Zero-Shot Machine-Generated Text Detection using Probability Curvaturecited by 726
  12. 2022GPT-NeoX-20B: An Open-Source Autoregressive Language Modelcited by 713
26 more
  1. 2023Benchmarking Large Language Models for News Summarizationcited by 445
  2. 2023Paraphrasing evades detectors of AI-generated text, but retrieval is an effective defensecited by 429
  3. 2023Can AI-Generated Text be Reliably Detected?cited by 386
  4. 2023RADAR: Robust AI-Text Detection via Adversarial Learningcited by 190
  5. 2016Winning Arguments: Interaction Dynamics and Persuasion Strategies in Good-faith Online Discussionscited by 110
  6. 2023MGTBench: Benchmarking Machine-Generated Text Detectioncited by 93
  7. 2022Same Author or Just Same Topic? Towards Content-Independent Style Representationscited by 86
  8. 2021Data Augmentation Approaches in Natural Language Processing: A Surveycited by 82
  9. 2023A Survey on Detection of LLMs-Generated Contentcited by 71
  10. 2023Fighting Fire with Fire: The Dual Role of LLMs in Crafting and Detecting Elusive Disinformationcited by 51
  11. 2022Attribution and Obfuscation of Neural Text Authorship: A Data Mining Perspectivecited by 41
  12. 2023CHEAT: A Large-scale Dataset for Detecting ChatGPT-writtEn AbsTractscited by 40
  13. 2023GPT-who: An Information Density-based Machine-Generated Text Detectorcited by 40
  14. 2022How Large Language Models are Transforming Machine-Paraphrased Plagiarismcited by 38
  15. 2019Explainable Authorship Verification in Social Media via Attention-based Similarity Learningcited by 37
  16. 2021ParaSCI: A Large Scientific Paraphrase Dataset for Longer Paraphrase Generationcited by 28
  17. 2022On the State of the Art in Authorship Attribution and Authorship Verificationcited by 27
  18. 2021Data Expansion using Back Translation and Paraphrasing for Hate Speech Detectioncited by 20
  19. 2021Does It Capture STEL? A Modular, Similarity-based Linguistic Style Evaluation Frameworkcited by 17
  20. 2023Can Authorship Representation Learning Capture Stylistic Features?cited by 17
  21. 2022Data Augmentation with Paraphrase Generation and Entity Extraction for Multimodal Dialogue Systemcited by 16
  22. 2021Learning to Reason for Text Generation from Scientific Tablescited by 14
  23. 2023Paraphrase Detection: Human vs. Machine Contentcited by 12
  24. 2023HANSEN: Human and AI Spoken Text Benchmark for Authorship Analysiscited by 10
  25. 2018Plagiarism: Taxonomy, Tools and Detection Techniquescited by 5
  26. 2019Style is NOT a single variable: Case Studies for Cross-Style Language Understandingcited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 8; 2025: 9; 2026: 6.

23 papers.

Sort:

By area:

  1. 2024Beemo: Benchmark of Expert-edited Machine-generated Outputscited by 27
  2. 2024CollabStory: Multi-LLM Collaborative Story Generation and Authorship Analysiscited by 26
  3. 2024Deconstructing The Ethics of Large Language Models from Long-standing Issues to New-emerging Dilemmas: A Surveycited by 23
  4. 2024Authorship Obfuscation in Multilingual Machine-Generated Text Detectioncited by 19
  5. 2024Effects of diversity incentives on sample diversity and downstream model performance in LLM-based text augmentationcited by 17
  6. 2024Spotting AI's Touch: Identifying LLM-Paraphrased Spans in Textcited by 16
  7. 2025Unveiling Attractor Cycles in Large Language Models: A Dynamical Systems View of Successive Paraphrasingcited by 15
  8. 2025HACo-Det: A Study Towards Fine-Grained Machine-Generated Text Detection under Human-AI Coauthoringcited by 15
  9. 2024MultiSocial: Multilingual Benchmark of Machine-Generated Text Detection of Social-Media Textscited by 14
  10. 2025Catch Me If You Can? Not Yet: LLMs Still Struggle to Imitate the Implicit Writing Styles of Everyday Authorscited by 10
  11. 2025OpenTuringBench: An Open-Model-based Benchmark and Framework for Machine-Generated Text Detection and Attributioncited by 9
  12. 2025Detection of LLM-Paraphrased Code and Identification of the Responsible LLM Using Coding Style Featurescited by 4
11 more
  1. 2025LLM in the Loop: Creating the ParaDeHate Dataset for Hate Speech Detoxificationcited by 4
  2. 2026Mutation Without Variation: Convergence Dynamics in LLM-Driven Program Evolutioncited by 4
  3. 2025Trends and Challenges in Authorship Analysis: A Review of ML, DL, and LLM Approachescited by 3
  4. 2026'Your AI Text is not Mine': Redefining and Evaluating AI-generated Text Detection under Realistic Assumptionscited by 3
  5. 2024Navigating the Shadows: Unveiling Effective Disturbances for Modern AI Content Detectorscited by 2
  6. 2025Integrating Personality into Digital Humans: A Review of LLM-Driven Approaches for Virtual Realitycited by 2
  7. 2025Personalized Author Obfuscation with Large Language Modelscited by 0
  8. 2026Headlines You Won't Forget: Can Pronoun Insertion Increase Memorability?cited by 0
  9. 2026Measuring, Localizing, and Ablating Alignment Signatures in LLMscited by 0
  10. 2026On the Indistinguishability of Human v/s AI Generated Textcited by 0
  11. 2026Style, Not Self: Surface Cues Explain Zero-Shot Code Attribution by Large Language Modelscited by 0