Open-Source LLMs for Vulnerability Classification Under Code Obfuscation: Efficiency Benchmarks
This report synthesises findings from 8 peer-reviewed papers addressing the following research question: How does the inference efficiency (latency, throughput) of Llama3, Codestral, and Deepseek R1 compare when classifying vulnerabilities in the Big-Vul dataset under increasing levels of code.
openaire +1 more source
A comprehensive bedside chest radiography dataset with structured, itemized and graded radiologic reports. [PDF]
Truhn D +8 more
europepmc +1 more source
Harnessing AI and analytics to enhance cybersecurity and privacy for collective intelligence systems. [PDF]
Naeem MR +6 more
europepmc +1 more source
MaskMyPy: python tools for performing and analyzing geographic masks. [PDF]
Swanlund D, Schuurman N.
europepmc +1 more source
SentinelSphere: An AI-driven cybersecurity platform integrating real-time threat detection with security awareness education. [PDF]
Tantaroudas ND +2 more
europepmc +1 more source
Stylometry for real-world expert coders: a zero-shot approach. [PDF]
Gurioli A, Gabbrielli M, Zacchiroli S.
europepmc +1 more source
Mitigating semantic label divergence in federated learning: Obfuscated encoding and alert filtering for security monitoring. [PDF]
Lee Y, Im J, Kim J, Yoon M.
europepmc +1 more source
Hybrid deep learning framework for accurate classification of high dimensional genomic data. [PDF]
Swain MK +5 more
europepmc +1 more source
OntoSecAI: Ontology-driven security automation for AI-enabled systems. [PDF]
Ullah U, Haleem M, Ullah A.
europepmc +1 more source
Design of an Automated Framework for Applying Generative AI-Based Source Code Obfuscation Techniques
Jihun Han +7 more
openaire +1 more source

