Research & Innovation

Global: Study Compares Prompt-Based and Fine-Tuned LLMs for Security Bug Report DetectionA research paper posted to arXiv on January 30, 2026, evaluates how large language models (LLMs) can be used…

Global: Adversarial Audio Perturbations Cause Transcription Errors and Identity Drift in ASR and Speaker VerificationA recent study published on arXiv in September 2025 examines how adversarial perturbations in speech affect…

Global: Performance Evaluation of Post-Quantum Cryptography in WPA‑Enterprise AuthenticationA recent academic study examined how integrating post‑quantum cryptography (PQC) into WPA‑Enterprise authentication impacts latency and security. Researchers built a testbed using…

Global: Residual Hybrid Quantum-Classical Model Improves Accuracy by Up to 55%A new hybrid quantum-classical architecture introduced by researchers from multiple institutions demonstrates significant performance gains while addressing the measurement bottleneck…

Global: New Assembly Flow Graph Method Enhances Explainability in Malware DetectionA team of computer scientists has introduced an Assembly Flow Graph (AFG) framework designed to improve transparency and performance in…

Global: Study Proposes Low-Recoverability Steganography in Fine-Tuned LLMs and Highlights Interpretability-Based DetectionResearchers from an unnamed institution have released a new preprint on arXiv that examines covert encoding of secret prompts…

Global: Secure NTT Architecture Detects Hardware Trojans and Side-Channel Attacks Researchers have introduced a new Number Theoretic Transform (NTT) design that can identify and mitigate both hardware Trojan‑induced control‑flow disruptions…

Global: Trackly Platform Offers Integrated User Behavior Analytics and Real-Time Anomaly DetectionIntegrated Analytics SolutionTrackly, a scalable software‑as‑a‑service offering, combines comprehensive user‑behavior analytics with rule‑based, real‑time anomaly detection to enhance digital…

Global: Researchers Introduce SilentWood Protocol to Speed Up Private Inference for Gradient Boosting ForestsResearchers have unveiled a new private inference protocol, named SilentWood, that leverages homomorphic encryption to enable faster,…

Global: Directed Greybox Fuzzing Expanded to Rust and Go Shows Superior Time-to-ExposureA research team has introduced a novel directed greybox fuzzing framework specifically engineered for Rust and Go applications, reporting…