Malicious Attachment

Malicious attachments, primarily delivered via email, represent a persistent cybersecurity threat. Current research focuses on improving detection methods, including the use of unsupervised topic modeling and advanced statistical analysis of file structures (like PDFs) to identify malicious code and deceptive email content. Furthermore, studies are investigating the vulnerability of large language models (LLMs) to generating or modifying malicious code, highlighting the need for robust defenses against LLM-assisted malware creation and obfuscation. These advancements aim to enhance threat detection and prevention, ultimately improving overall cybersecurity.

Papers

September 23, 2024

RMCBench: Benchmarking Large Language Models' Resistance to Malicious Code
Jiachi Chen, Qingyuan Zhong, Yanlin Wang, Kaiwen Ning, Yongkun Liu, Zenan Xu, Zhe Zhao, Ting Chen, Zibin Zheng
Effective Resistance Malicious Code Malicious Attachment

July 11, 2024

Uncovering Semantics and Topics Utilized by Threat Actors to Deliver Malicious Attachments and URLs
Andrey Yakymovych, Abhishek Singh
Malicious Agent Significant Topic Threat Detection Semantic Cue Uniform Resource Locator Malicious Attachment

June 27, 2024

Synthetic Cancer -- Augmenting Worms with LLMs
Benjamin Zimmerman, David Zollikofer
Large Language Model Malware Attack Advanced Large Language Model Synthetic Tumor Metamorphic Malware Malicious Attachment

November 8, 2021

HAPSSA: Holistic Approach to PDF Malware Detection Using Signal and Statistical Analysis
Tajuddin Manhar Mohammed, Lakshmanan Nataraj, Satish Chikkagoudar, Shivkumar Chandrasekaran, B. S. Manjunath
Malware Detection Abnormal Signal Holistic Approach Malware Analysis Statistical Analysis Novel Malware Malware Sample Malicious Attachment

Malicious Attachment

Papers

RMCBench: Benchmarking Large Language Models' Resistance to Malicious Code

Uncovering Semantics and Topics Utilized by Threat Actors to Deliver Malicious Attachments and URLs

Synthetic Cancer -- Augmenting Worms with LLMs

HAPSSA: Holistic Approach to PDF Malware Detection Using Signal and Statistical Analysis