Microsoft

Inside Tycoon2FA: How a leading AiTM phishing kit operated at scale

Tycoon2FA has become a leading phishing-as-a-service (PhaaS) platforms, enabling campaigns that reach over 500,000 ...
IEEE

Mask R-CNN with HRNet Backbone for Person Detection Evaluation on COCO Using mIoU and Threshold Optimization

Abstract: A crucial computer vision problem is person detection, but real-world challenges including crowded backgrounds, inconsistent lighting, and object occlusion require accurate and robust models ...
GitHub

uqlm: Uncertainty Quantification for Language Models

UQLM provides a suite of response-level scorers for quantifying the uncertainty of Large Language Model (LLM) outputs. Each scorer returns a confidence score between 0 and 1, where higher scores ...
IEEE

Binary Code Analysis for Cybersecurity: A Systematic Review of Forensic Techniques in Vulnerability Detection and Anti-Evasion Strategies

Abstract: Binary code analysis is essential in modern cybersecurity, examining compiled program outputs to identify vulnerabilities, detect malware, and ensure software security compliance. However, ...
GitHub

DeFB: Decomposed Feature Learning for Real-Time Multi-Person Eyeblink Detection in Untrimmed In-the-Wild Videos

🔍 Rethinking Unified Models: We identify two critical limitations in existing unified multi-person eyeblink detection models: (1) feature granularity conflict between face localization and eyeblink ...
The Hacker News

Microsoft Develops Scanner to Detect Backdoors in Open-Weight Large Language Models

Microsoft on Wednesday said it built a lightweight scanner that it said can detect backdoors in open-weight large language models (LLMs) and improve the overall trust in artificial intelligence (AI) ...