Mitre ATT&CK
PulseAugur coverage of Mitre ATT&CK — every cluster mentioning Mitre ATT&CK across labs, papers, and developer communities, ranked by signal.
-
LLM agents vulnerable to Morse code and other encoding attacks
Security researchers demonstrated a novel prompt injection attack against Bankr, an AI financial assistant, by encoding instructions in Morse code. This method bypassed traditional content filters because the LLM interp…
-
LLMs leverage code analysis for improved malware attribution
Researchers have developed LCC-LLM, a framework and dataset designed to improve malware attribution using large language models. The system leverages code-centric representations, including decompiled C code and assembl…
-
CyberAId platform uses AI agents to bolster financial cybersecurity
A new paper proposes CyberAId, a hybrid multi-agent system designed to enhance cybersecurity for financial institutions. The system integrates specialized AI sub-agents with existing SIEM/XDR telemetry, rather than repl…
-
Retrieval-Augmented LLMs Enhance Cybersecurity Incident Analysis Efficiency
Researchers have developed a Retrieval-Augmented Generation (RAG) system to automate the analysis of cybersecurity incidents. This system uses targeted queries and a library of MITRE ATT&CK techniques to extract indicat…
-
DeepStage uses AI to learn autonomous defense against multi-stage cyberattacks
Researchers have developed DeepStage, a new framework utilizing deep reinforcement learning to create autonomous defense policies against multi-stage cyberattacks. The system models enterprise environments as partially …
-
OntoLogX uses LLMs to extract actionable threat intelligence from cybersecurity logs
Researchers have developed OntoLogX, an AI agent designed to extract Cyber Threat Intelligence (CTI) from raw cybersecurity logs. The system utilizes Large Language Models (LLMs) combined with a lightweight log ontology…
-
New AI tool automates threat modeling for cyber-physical systems
Researchers have developed SMSI, a novel pipeline that automates threat modeling for cyber-physical systems. This system translates architectural models into actionable security control recommendations by mapping system…
-
Researchers develop self-supervised learning for Android malware detection
Researchers have developed a new method for detecting Android malware that addresses temporal bias in machine learning models. By constructing a time-stamped dataset and implementing a timestamp-verification procedure, …
-
TraceScope system automates phishing URL triage with AI agents
Researchers have developed TraceScope, an interactive system designed to combat sophisticated phishing campaigns that evade traditional URL classifiers. The system uses a sandboxed agent to navigate potentially maliciou…