Saved in:
Bibliographic Details
Main Authors: Shokri, Mohammad, Levitan, Sarah Ita, Levitan, Rivka
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.12090
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910951074168832
author Shokri, Mohammad
Levitan, Sarah Ita
Levitan, Rivka
author_facet Shokri, Mohammad
Levitan, Sarah Ita
Levitan, Rivka
contents In this paper, we investigate the efficacy of large language models (LLMs) in obfuscating authorship by paraphrasing and altering writing styles. Rather than adopting a holistic approach that evaluates performance across the entire dataset, we focus on user-wise performance to analyze how obfuscation effectiveness varies across individual authors. While LLMs are generally effective, we observe a bimodal distribution of efficacy, with performance varying significantly across users. To address this, we propose a personalized prompting method that outperforms standard prompting techniques and partially mitigates the bimodality issue.
format Preprint
id arxiv_https___arxiv_org_abs_2505_12090
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Personalized Author Obfuscation with Large Language Models
Shokri, Mohammad
Levitan, Sarah Ita
Levitan, Rivka
Computation and Language
Artificial Intelligence
In this paper, we investigate the efficacy of large language models (LLMs) in obfuscating authorship by paraphrasing and altering writing styles. Rather than adopting a holistic approach that evaluates performance across the entire dataset, we focus on user-wise performance to analyze how obfuscation effectiveness varies across individual authors. While LLMs are generally effective, we observe a bimodal distribution of efficacy, with performance varying significantly across users. To address this, we propose a personalized prompting method that outperforms standard prompting techniques and partially mitigates the bimodality issue.
title Personalized Author Obfuscation with Large Language Models
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2505.12090