Research & Innovation

Global: Study Benchmarks Efficient Implementations of Differentially Private SGDA team of machine learning researchers released new benchmark results in June 2024 to assess the computational cost of training deep learning…

Global: Researchers Achieve Near‑Perfect Accuracy Classifying Post‑Quantum Cryptography ImplementationsA recent arXiv preprint released in March 2025 reports that machine‑learning‑based classification techniques can reliably identify post‑quantum (PQ) cryptographic algorithms across a range…

Global: New Architecture Merges QES-Grade PKCS#11 Keys with Virtual FIDO2 AuthenticatorsResearchers have introduced a design that combines qualified electronic signature (QES) tokens with the FIDO2/WebAuthn authentication framework, aiming to provide…

Global: Researchers Identify 24 Factors Influencing Belief in MisinformationResearchers have compiled a comprehensive literature review that examines why individuals accept false information, including misinformation, disinformation, and fabricated content. The study,…

Global: Study Evaluates Zero Trust Architecture Viability for Small‑Medium BusinessesResearchers have introduced an integrated predictive model that gauges both the feasibility of adopting Zero Trust Architecture (ZTA) and its effectiveness…

Global: Simple Power Analysis Attack on HQC Demonstrated with High Success RateOn Jan 12 2026, researchers Pavel Velek, Tomáš Rabas, and Jiří Buček published a preprint on arXiv describing a side‑channel attack…

Global: Implicit Tool Poisoning Framework Demonstrates High Success Rate in LLM Agent AttacksResearchers from an unnamed institution have introduced MCP-ITP, an automated framework designed to embed malicious instructions in tool…

Global: ViPer Framework Achieves Near-Human Success Against Major Visual Reasoning CAPTCHAsResearchers introduced ViPer, a unified attack framework for Visual Reasoning CAPTCHAs (VRCs), in a paper posted to arXiv in January…

Global: Leveraging Membership Inference Attacks for Privacy Measurement in Federated Learning for Remote Sensing Images On January 8, 2026, researchers Anh‑Kiet Duong, Petra Gomez‑Krämer, Hoàng‑Ân Lê and Minh‑Tan Pham posted…

Global: Evaluation of LLMs for Secure PowerShell Script GenerationResearchers from an unnamed institution released a study on January 2026 that introduces a new benchmark, SecGenEval-PS, to assess large language models’…