News
Context Based Password Cracking Dictionary Expansion Using Generative Pre-trained Transformers
Congratulations to Greta Imhof and co-authors Aikaterini Kanta, and Mark Scanlon on the publication of Context Based Password Cracking Dictionary Expansion Using Generative Pre-trained Transformers in 2024 Cyber Research Conference - Ireland (Cyber-RCI).
Co-authors: Aikaterini Kanta, and Mark Scanlon.
AI-generated summary of the contribution: This research focuses on modern password cracking techniques, specifically context-based password cracking and the use of large language models. The authors create context-based password dictionaries by training PassGPT models with contextual information, such as demographics, interests, and hobbies. The effectiveness of this approach is evaluated on ten datasets of leaked passwords from various topics, showing improved password cracking efficiency and accuracy compared to traditional methods. The study contributes to the development of more efficient and effective password cracking tools for law enforcement agencies in digital forensic investigations.