Conference
Combining Variational Autoencoders and Transformer Language Models for Improved Password Generation
| العنوان: | Combining Variational Autoencoders and Transformer Language Models for Improved Password Generation |
|---|---|
| المؤلفون: | Biesner, David, Cvejoski, Kostadin, Sifa, Rafet |
| سنة النشر: | 2022 |
| المجموعة: | Publikationsdatenbank der Fraunhofer-Gesellschaft |
| مصطلحات موضوعية: | language models, latent variable models, neural networks, passwords, text generation, transformers |
| الوصف: | Password generation techniques have recently been explored by leveraging deep-learning natural language processing (NLP) algorithms. Previous work has raised the state of the art for password guessing algorithms significantly, by approaching the problem using either variational autoencoders with CNN-based encoder and decoder architectures or transformer-based architectures (namely GPT2) for text generation. In this work we aim to combine both paradigms, introducing a novel architecture that leverages the expressive power of transformers with the natural sampling approach to text generation of variational autoencoders. We show how our architecture generates state-of-the-art results in password matching performance across multiple benchmark datasets. |
| نوع الوثيقة: | conference object |
| وصف الملف: | application/pdf |
| اللغة: | English |
| Relation: | International Conference on Availability, Reliability and Security 2022; #PLACEHOLDER_PARENT_METADATA_VALUE#; Proceedings of the 17th International Conference on Availability, Reliability and Security, ARES 2022; ML2R; 1%7CS18038B, 01%7CS18038C; https://publica.fraunhofer.de/handle/publica/429936; https://doi.org/10.24406/publica-632 |
| DOI: | 10.1145/3538969.3539000 |
| DOI: | 10.24406/publica-632 |
| الاتاحة: | https://publica.fraunhofer.de/handle/publica/429936 https://doi.org/10.1145/3538969.3539000 https://doi.org/10.24406/publica-632 |
| Rights: | open access ; CC BY 4.0 |
| رقم الانضمام: | edsbas.377AC3F1 |
| قاعدة البيانات: | BASE |
| DOI: | 10.1145/3538969.3539000 |
|---|