Skip to main content

Are You Robert or RoBERTa? Deceiving Online Authorship Attribution Models Using Neural Text Generators

Jones, Keenan, Nurse, Jason R. C., Li, Shujun (2022) Are You Robert or RoBERTa? Deceiving Online Authorship Attribution Models Using Neural Text Generators. In: 16th International AAAI Conference on Web and Social Media (ICWSM-22), 6-9 June 2022, Atlanta, USA. (KAR id:93628)

PDF Publisher pdf
Language: English
Download (389kB) Preview
[thumbnail of 19304-Article Text-23317-1-2-20220531.pdf]
Preview
This file may not be suitable for users of assistive technology.
Request an accessible format

Abstract

Recently, there has been a rise in the development of powerful pre-trained natural language models, including GPT-2, Grover, and XLM. These models have shown state-of-the-art capabilities towards a variety of different NLP tasks, including question answering, content summarisation, and text generation. Alongside this, there have been many studies focused on online authorship attribution (AA). That is, the use of models to identify the authors of online texts. Given the power of natural language models in generating convincing texts, this paper examines the degree to which these language models can generate texts capable of deceiving online AA models. Experimenting with both blog and Twitter data, we utilise GPT-2 language models to generate texts using the existing posts of online users. We then examine whether these AI-based text generators are capable of mimicking authorial style to such a degree that they can deceive typical AA models. From this, we find that current AI-based text generators are able to successfully mimic authorship, showing capabilities towards this on both datasets. Our findings, in turn, highlight the current capacity of powerful natural language models to generate original online posts capable of mimicking authorial style sufficiently to deceive popular AA methods; a key finding given the proposed role of AA in real world applications such as spam-detection and forensic investigation.

Item Type: Conference or workshop item (Paper)
Uncontrolled keywords: Authorship Attribution, Authorship Analysis, Natural Language Generation, NLG, GPT-2, Natural Language Processing, NLP, Deception, Twitter, Blog, Language Models, Cybercrime, Social Media
Subjects: Q Science > QA Mathematics (inc Computing science) > QA 76 Software, computer programming,
Q Science > QA Mathematics (inc Computing science) > QA 76 Software, computer programming, > QA76.87 Neural computers, neural networks
T Technology
Divisions: Divisions > Division of Computing, Engineering and Mathematical Sciences > School of Computing
Depositing User: Keenan Jones
Date Deposited: 17 Mar 2022 13:52 UTC
Last Modified: 07 Jun 2022 10:31 UTC
Resource URI: https://kar.kent.ac.uk/id/eprint/93628 (The current URI for this page, for reference purposes)
Jones, Keenan: https://orcid.org/0000-0001-9042-1308
Nurse, Jason R. C.: https://orcid.org/0000-0003-4118-1680
Li, Shujun: https://orcid.org/0000-0001-5628-7328
  • Depositors only (login required):

Downloads

Downloads per month over past year