Two Birds with One Stone: Multi-Task Detection and Attribution of LLM-Generated Text

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Rao, Zixin, Mohamed, Youssef, Liu, Shang, Liu, Zeyan
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866912545044955136
author Rao, Zixin
Mohamed, Youssef
Liu, Shang
Liu, Zeyan
author_facet Rao, Zixin
Mohamed, Youssef
Liu, Shang
Liu, Zeyan
contents Large Language Models (LLMs), such as GPT-4 and Llama, have demonstrated remarkable abilities in generating natural language. However, they also pose security and integrity challenges. Existing countermeasures primarily focus on distinguishing AI-generated content from human-written text, with most solutions tailored for English. Meanwhile, authorship attribution--determining which specific LLM produced a given text--has received comparatively little attention despite its importance in forensic analysis. In this paper, we present DA-MTL, a multi-task learning framework that simultaneously addresses both text detection and authorship attribution. We evaluate DA-MTL on nine datasets and four backbone models, demonstrating its strong performance across multiple languages and LLM sources. Our framework captures each task's unique characteristics and shares insights between them, which boosts performance in both tasks. Additionally, we conduct a thorough analysis of cross-modal and cross-lingual patterns and assess the framework's robustness against adversarial obfuscation techniques. Our findings offer valuable insights into LLM behavior and the generalization of both detection and authorship attribution.
format Preprint
id arxiv_https___arxiv_org_abs_2508_14190
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Two Birds with One Stone: Multi-Task Detection and Attribution of LLM-Generated Text
Rao, Zixin
Mohamed, Youssef
Liu, Shang
Liu, Zeyan
Cryptography and Security
Computation and Language
Machine Learning
Large Language Models (LLMs), such as GPT-4 and Llama, have demonstrated remarkable abilities in generating natural language. However, they also pose security and integrity challenges. Existing countermeasures primarily focus on distinguishing AI-generated content from human-written text, with most solutions tailored for English. Meanwhile, authorship attribution--determining which specific LLM produced a given text--has received comparatively little attention despite its importance in forensic analysis. In this paper, we present DA-MTL, a multi-task learning framework that simultaneously addresses both text detection and authorship attribution. We evaluate DA-MTL on nine datasets and four backbone models, demonstrating its strong performance across multiple languages and LLM sources. Our framework captures each task's unique characteristics and shares insights between them, which boosts performance in both tasks. Additionally, we conduct a thorough analysis of cross-modal and cross-lingual patterns and assess the framework's robustness against adversarial obfuscation techniques. Our findings offer valuable insights into LLM behavior and the generalization of both detection and authorship attribution.
title Two Birds with One Stone: Multi-Task Detection and Attribution of LLM-Generated Text
topic Cryptography and Security
Computation and Language
Machine Learning
url https://arxiv.org/abs/2508.14190