
LeakGauge
Detects LLM context-leakage attacks by training lightweight behavior probes on log-probabilities, with vLLM offline/server detection pipelines.

Detects LLM context-leakage attacks by training lightweight behavior probes on log-probabilities, with vLLM offline/server detection pipelines.

Jailbreak vulnerability across early-exit depths LLMs. Published at NeurIPS 2026 in the FLMSec workshop

ML-based detection of Zombie ZIP archive header evasion attacks (CVE-2026-0866)

An implementation of the FAIR CRQ Framework

Prohibited Items Segmentation via Occlusion-aware Bilayer Modeling (ICME 2025)

A novel adversarial attack on LLM based on the Exponentiated Gradient Descent technique.

CLPSTNet: A Progressive Multi-Scale Convolutional Steganography Model Integrating Curriculum Learning

Adversarial image attacks on vision-language web agents, from visual grounding to browser execution

Public version of the Entropica repo

détection des attaques sql/xss sur API web avec IA

This repository hosts a multimodal web attack dataset (MWAD) to advance AI-driven threat detection research.

Production AI defense with 7-layer protection: mathematical constraints, object-capability access, distributed O2 consensus, SVETILO ethics. First…

AGLS

Code for ACL 2026 (main) paper "DeepGuard: Secure Code Generation via Multi-Layer Semantic Aggregation"

A Red Team vs. Blue Team Adversarial AI Simulation.

Research code for poisoning attacks on the PGM-index, demonstrating how to craft adversarial data to degrade learned index performance.

The Governed Agentic AI Operating System — Rust + Tauri 2.0 | 65 crates, 658 commands, 84 pages, 5,029 tests, 10/10 OWASP

VulnAgent-X: A Layered Agentic Framework for Repository-Level Vulnerability Detection