PagPassGPT: Pattern Guided Password Guessing via Generative Pretrained Transformer

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Su, Xingyu, Zhu, Xiaojie, Li, Yang, Li, Yong, Chen, Chi, Esteves-Veríssimo, Paulo
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866911921452613632
author Su, Xingyu
Zhu, Xiaojie
Li, Yang
Li, Yong
Chen, Chi
Esteves-Veríssimo, Paulo
author_facet Su, Xingyu
Zhu, Xiaojie
Li, Yang
Li, Yong
Chen, Chi
Esteves-Veríssimo, Paulo
contents Amidst the surge in deep learning-based password guessing models, challenges of generating high-quality passwords and reducing duplicate passwords persist. To address these challenges, we present PagPassGPT, a password guessing model constructed on Generative Pretrained Transformer (GPT). It can perform pattern guided guessing by incorporating pattern structure information as background knowledge, resulting in a significant increase in the hit rate. Furthermore, we propose D&C-GEN to reduce the repeat rate of generated passwords, which adopts the concept of a divide-and-conquer approach. The primary task of guessing passwords is recursively divided into non-overlapping subtasks. Each subtask inherits the knowledge from the parent task and predicts succeeding tokens. In comparison to the state-of-the-art model, our proposed scheme exhibits the capability to correctly guess 12% more passwords while producing 25% fewer duplicates.
format Preprint
id arxiv_https___arxiv_org_abs_2404_04886
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle PagPassGPT: Pattern Guided Password Guessing via Generative Pretrained Transformer
Su, Xingyu
Zhu, Xiaojie
Li, Yang
Li, Yong
Chen, Chi
Esteves-Veríssimo, Paulo
Cryptography and Security
Artificial Intelligence
Amidst the surge in deep learning-based password guessing models, challenges of generating high-quality passwords and reducing duplicate passwords persist. To address these challenges, we present PagPassGPT, a password guessing model constructed on Generative Pretrained Transformer (GPT). It can perform pattern guided guessing by incorporating pattern structure information as background knowledge, resulting in a significant increase in the hit rate. Furthermore, we propose D&C-GEN to reduce the repeat rate of generated passwords, which adopts the concept of a divide-and-conquer approach. The primary task of guessing passwords is recursively divided into non-overlapping subtasks. Each subtask inherits the knowledge from the parent task and predicts succeeding tokens. In comparison to the state-of-the-art model, our proposed scheme exhibits the capability to correctly guess 12% more passwords while producing 25% fewer duplicates.
title PagPassGPT: Pattern Guided Password Guessing via Generative Pretrained Transformer
topic Cryptography and Security
Artificial Intelligence
url https://arxiv.org/abs/2404.04886