Topic
automated
MASCOT-Android: Automated Pipeline and Curated Dataset for Android Malware Source Code Discovery
Researchers introduce MASCOT-Android, a curated dataset of Android malware source code and an automated collection framework. The key finding is that repository-level documentation alone provides a strong signal, enabling a LinearSVC classifier to achieve 96.28% accuracy with a 1.06% false positive rate. The model outputs confidence scores for threshold adjustment, making it practical for real-world malware source code collection.
New Automated Jailbreak Attack UNIATTACK Achieves High Success Rate Against Multi-Layered LLM Defenses
Researchers present UNIATTACK, an adversarial testing framework that extracts high-impact attack features from existing exploits and uses a specialized attacker LLM to compose flexible templates. The framework achieves an average attack success rate improvement of 64.63% to 248.82% over baselines on models with multi-layered defenses, while costing only 0.03% to 4.96% of baseline costs.
AIChilles Automatically Unearths Hidden Weaknesses in AI-Evolved Programs
Researchers developed AIChilles, an automated tool that uncovers hidden weaknesses in AI-evolved programs. Testing 30 AI-generated programs across five system applications, it found 49 distinct failures in correctness, runtime, memory, and output quality. The tool combines workload extraction, constraint inference, and differential oracles to identify regressions that could undermine AI-generated code reliability.