CAMLIS 2023
Conference on Applied Machine Learning in Information Security

Table of Contents

• Web Content Filtering Through Knowledge Distillation of Large Language Models 1-21
  Tamás Vörös, Sean Bergeron, Konstantin Berlin
• Defending the unknown: Exploring reinforcement learning agents’ deployment in realistic, unseen networks 22-35
  Alberto Acuto, Simon Maskell, Jack D.
• Multi-Agent Reinforcement Learning for Maritime Operational Technology Cyber Security 36-48
  Alec Wilson, Ryan Menzies, Neela Morarji, David Foster, Marco Casassa Mont, Esin Turkbeyler, Lisa Gralewski
• Detecting SQL Injection Attacks using Machine Learning 49-59
  Béatrice Moissinac, Elie Saad, Miranda Clay, Maialen Berrondo
• Anomaly Detection of Command Shell Sessions based on DistilBERT: Unsupervised and Supervised Approaches 60-76
  Zefang Liu, John F. Buford
• Enhancing Exfiltration Path Analysis Using Reinforcement Learning 77-90
  Riddam Rishu, Akshay Kakka, Cheng Wang, Abdul Rahman, Christopher Redino, Dhruv Nandakumar, Tyler Cody, Ryan Clark, Daniel Radke, Edward Bowen
• Model Leeching: An Extraction Attack Targeting LLMs 91-104
  Lewis Birch, William Hackett, Stefan Trawicki, Neeraj Suri, Peter Garraghan
• MalDICT: Benchmark Datasets on Malware Behaviors, Platforms, Exploitation, and Packers 105-121
  Robert J. Joyce, Edward Raff, Charles Nicholas, James Holt
• Proxy in a Haystack: Uncovering and Classifying MFA Bypass Phishing Attacks in Large-Scale Authentication Data 122-133
  Rebecca Lynch, Lauren Saue-Fletcher
• Adaptive Experimental Design for Intrusion Data Collection 134-151
  Kate Highnam, Zach Hanif, Ellie Van Vogt, Sonali Parbhoo, Sergio Maffeis, Nicholas R. Jennings
• Half-Day Vulnerabilities: A study of the First Days of CVE Entries 152-180
  Kobra Khanmohammadi, Raphaël Khoury
• Small Effect Sizes in Malware Detection? Make Harder Train/Test Splits! 181-192
  Tirth Patel, Fred Lu, Edward Raff, Charles Nicholas, Cynthia Matuszek, James Holt
• FASER: Binary Code Similarity Search through the use of Intermediate Representations 193-202
  Josh Collyer, Tim Watson, Iain Phillips