metta

Metta Metta es una herramienta de preparación para la seguridad de la información. Este proyecto utiliza Redis/Celery, Python y Vagrant con VirtualBox para realizar simulación adversarial. Esto le permite probar principalmente su instrumentación basada en el host, pero también puede permitirle...

deep-pwning

Deep-pwning è un framework leggero per sperimentare con modelli di machine learning, con l'obiettivo di valutarne la robustezza contro un avversario motivato. Si noti che deep-pwning nel suo stato attuale è lontano dall'essere maturo o completo. È pensato per essere sperimentato, ampliato ed este...

foolbox

home: true heroImage: /logo.png heroText: Foolbox tagline: "Foolbox: attacchi adversarial veloci per valutare la robustezza dei modelli di machine learning in PyTorch, TensorFlow e JAX" actionText: Inizia → actionLink: /guide/ features: title: Prestazioni native details: Foolbox 3 è basata su...

Online AutoML: Evaluating Poisoning Attacks on Adversarial Training Defense Strategy in IoT Networks

Machine learning ML-powered poisoning attack vectors are adversarial maneuvers whereby an attacker intentionally inserts, corrupts, or alters training data to distort an ML model's learning process. The objective is to diminish model efficacy, instill biases, induce misclassifications, or include...

Adaptive Model Inversion Attacks Generalize a Privacy-Robustness Tradeoff

In this paper, we show that standard evaluations of high-resolution Model Inversion Attacks MIAs significantly underestimate training-data privacy leakage. State-of-the-art privacy defenses, standard training techniques such as MixUp and Adversarial Training, and undefended models all leak traini...

COPEX: Benchmarking LLM Robustness to Adversarial Context across Model Context Protocol Layers

Large language models increasingly mediate tool use in Model Context Protocol MCP systems, where adversarial influence may enter through user instructions, tool schemas, tool outputs, or protocol messages. Existing benchmarks often evaluate deployed agents, conflating model susceptibility with...

JASPER: Special Session on Joint Reliability and Security Assessment of SPlit Computing for Edge Robustness

Split Computing SC enables efficient deployment of Deep Neural Networks DNNs by partitioning inference between edge devices and cloud servers. However, intermediate feature representations are simultaneously exposed to hardware faults and adversarial attacks, which are traditionally evaluated...

EvoRiskBench: An Evolving Benchmark for Runtime Security Risks in Workspace Agents

Workspace agents combine large language models with execution harnesses to perform stateful, multi-step tasks that access or modify external resources. Existing benchmarks leave gaps in executable coverage of their runtime security risks, while evolving model capabilities, harnesses, tools, and...

OpenAI Disrupts Reasoning Extraction Campaign Linked to Moonshot AI Associates

OpenAI on Wednesday said it identified and disrupted a coordinated distillation campaign that was designed to illicitly extract protected reasoning from its artificial intelligence AI models. A "core cluster of the activity," going back to the first week of July, has been attributed to individual...

ac-solver (0.1.0), acedeploy (>=2.4.15, <=2.4.366.dev1) +778 more potentially affected by CVE-2026-87817 via gitpython (>=3.0.0, <=3.1.59)

gitpython PYPI version =3.0.0, =2.4.15, =2025.10.17, =0.4.0, =0.4.0, =1.2.3, =0.4.7, =0.4.7, =0.2.0, =0.1.8, =0.87.2.dev9, =0.5.0, =0.1.0.dev19, =1.3.0, =2.0.3 and more Source cves: CVE-2026-87817 Source advisory: GHSA-239G-WHFQ-7XJ9...

Exploiting Vulnerabilities: Universal Adversarial Attacks on Vision-Language-Action Models in Robotics

Recently, Vision-Language-Action VLA models have revolutionized robotic manipulation by seamlessly integrating visual perception, language understanding, and action generation in an end-to-end learning framework. However, since these models are designed to interact directly with the physical worl...

MADBench: Benchmarking the Security of Multi-Agent Debate

Multi-agent debate MAD can improve large language model LLM reasoning by allowing multiple agents to exchange and critique their answers to the same task. However, the interactions that enable agents to correct mistakes can also spread adversarial errors and steer the agents toward an incorrect...

Memetic Trojans: Social Contagions As Carriers of Adversarial Payloads in Agent Networks

Autonomous large language model LLM agents increasingly interact in network environments where adversarial content can propagate between agents. Known attacks include agent worms, which spread through self-replicating prompt injections or configuration compromises. We introduce memetic trojans, a...

SURE: Framework for Safety to Construct Trustworthy AI

Warning: This paper contains harmful and offensive text. Recently, large language models such as GPT-4, and Claude have revolutionized tasks in various domains. As the use of these large language models increases, people are increasingly concerned about AI safety and demand that large language...

LogiC-Diff: Embedding Security Properties into AI-Enabled Cyber-Physical Systems

AI-enabled Cyber-Physical Systems CPS are highly vulnerable to adversarial and anomalous inputs, where small perturbations can induce cascading errors and unsafe control actions. Existing approaches, such as rule-based filtering, training-time regularization, or diffusion-based reconstruction,...

Inference-Layer Security: Defending against Adversarial Inference and Infrastructure Abuse

A Technical Report: Operating a large language model LLM as a service requires more than inference infrastructure: the provider must also defend against adversarial interactions that seek to exploit the service, including jailbreaking for harmful use, sophisticated denial of service, and...

Breaking Windows Malware Detection: A Comprehensive Evaluation of Problem-Space Adversarial Robustness

Problem-space evasion attacks have exposed critical weaknesses in machine learning-based malware detectors; yet, their evaluation remains fragmented across models, datasets, and attack methodologies, often neglecting domain-specific requirements such as executability and functionality preservatio...

FinRT: Distilling Adaptive Red-Teaming Strategies into Reusable Adversarial Generators in Consumer Finance

In regulated industries like consumer finance, seemingly harmless user queries can exploit large language model vulnerabilities, triggering safety failures and pushing responses dangerously close to policy limits. Existing automated red-teaming methods trade off attack effectiveness against...

Adversarial Debiasing of Machine Learning Models for Enhanced Network Security against DDoS Attacks

Distributed Denial of Service attacks are a growing threat to network infrastructure, and new techniques, including the use of generative AI, make them harder to detect. Traditional detection systems, such as rule based firewalls, often fail to identify these evolving attack patterns. In this...

ActBench

ActBench ActBench is a self-evolving benchmark of behavioral safety in cowork agents. It defines behavioral safety as whether an agent's execution remains within the permissions and state changes required by a benign task, and evaluates realized behavioral risk from execution trajectories rather...