PISmith

PISmith : Red Teaming basé sur l'apprentissage par renforcement pour les défenses contre l'injection de prompts COLM 2026 Ceci est une implémentation officielle de PISmith : Red Teaming basé sur l'apprentissage par renforcement pour les défenses contre l'injection de prompts Configuration de...

GuardReasoner-VL

GuardReasoner-VL: Sécuriser les VLM grâce au raisonnement renforcé Yue Liu, Shengfang Zhai, Mingzhe Du Yulin Chen, Tri Cao, Hongcheng Gao, Cheng Wang Xinfeng Li, Kun Wang, Junfeng Fang, Jiaheng Zhang, Bryan Hooi 1National University of Singapore, 2Nanyang Technological University Pour renforcer l...

StealthRL

StealthRL : Attaques de paraphrase par apprentissage par renforcement pour l'évasion multi-détecteurs des détecteurs de texte IA Article arXiv Démo Modèle Hugging Face Jeu de données de référence Hugging Face Résumé Les détecteurs de texte IA sont de plus en plus utilisés dans des contextes à for...

CyberBattleSim

CyberBattleSim 8 أبريل 2021: انظر الإعلان على مدونة أمن مايكروسوفت. CyberBattleSim هي منصة بحث تجريبية للتحقيق في تفاعل العوامل الآلية التي تعمل في بيئة شبكة مؤسسية محاكاة مجردة. توفر المحاكاة تجريدًا عالي المستوى لشبكات الحاسوب ومفاهيم الأمن السيبراني. تسمح واجهة OpenAI Gym المبنية على بايثون...

AutoPentest-DRL

AutoPentest-DRL: اختبار الاختراق الآلي باستخدام التعلم المعزز العميق AutoPentest-DRL هو إطار عمل لاختبار الاختراق الآلي يعتمد على تقنيات التعلم المعزز العميق DRL. يمكن لـ AutoPentest-DRL تحديد مسار الهجوم الأكثر ملاءمة لشبكة منطقية معينة، ويمكن استخدامه أيضًا لتنفيذ هجوم اختبار اختراق على شبكة...

Pesidious

تحوير البرمجيات الخبيثة باستخدام التعلم المعزز العميق والشبكات التنافسية التوليدية الغرض من هذه الأداة هو استخدام الذكاء الاصطناعي لتحوير عينة من البرمجيات الخبيثة PE32 فقط لتجاوز المصنفات المدعومة بالذكاء الاصطناعي مع الحفاظ على وظائفها سليمة. في الماضي، تم إجراء أعمال بارزة في هذا المجال، حيث ن...

OpenAI Pauses Frontier RL Training as It Tightens Defenses Against Unsafe AI Behavior

OpenAI on Tuesday revealed that it paused reinforcement learning RL training for its latest artificial intelligence AI models for two weeks while it shored up additional defenses and increased the scope of its monitoring to avert another Hugging Face-like incident. "As models become more capable,...

ARTA: Adaptive Reinforcement-Learning-Based Throttling Agent for RowHammer Vulnerabilities

RowHammer vulnerability continues to intensify with DRAM scaling, reducing the activation threshold needed to induce bitflips and rendering existing defenses such as TRR, ECC, and refresh-based mechanisms vulnerable to sophisticated multi-bank hammering patterns. This work presents ARTA, a...

A Red Teaming Framework for Evaluating Robustness of AI-Enabled Security Orchestration, Automation, and Response Systems

AI-enabled Security Orchestration, Automation, and Response SOAR systems increasingly employ autonomous agents for cyber defense, yet their resilience to adaptive adversaries is underexplored. We introduce an autonomous red teaming framework that integrates large language models LLMs with...

Operationalizing Cybersecurity Governance for Mitigation Planning with Attack-Path Modeling and Reinforcement Learning

We address a fundamental challenge in cybersecurity operations of translating governance frameworks into actionable mitigation decisions under realistic resource constraints. Frameworks such as the NIST Cybersecurity Framework CSF provide widely adopted measures of organizational maturity, but do...

STARE: Step-Wise Temporal Alignment and Red-Teaming Engine for Multi-Modal Toxicity Attack

Red-teaming Vision-Language Models is essential for identifying vulnerabilities where adversarial image-text inputs trigger toxic outputs. Existing approaches treat image generation as a black box, returning only terminal toxicity scores and leaving open the question of when and how toxic semanti...

XekRung Technical Report

We present XekRung, a frontier large language model for cybersecurity, designed to provide comprehensive security capabilities. To achieve this, we develop diverse data synthesis pipelines tailored to the cybersecurity domain, enabling the scalable construction of high-quality training data and...

Risk Models As Mediating Artifacts: A Postphenomenological Analysis of the CIIM Framework in Cybersecurity Practice

This article applies postphenomenological theory to the field of cybersecurity risk management, arguing that formal risk models function as mediating artifacts that shape how security practitioners or analysts perceive, interpret, and act on threats. Based on Don Ihde's taxonomy on human-technolo...

VeRL 权限许可和访问控制问题漏洞

VeRL is an open-source reinforcement learning framework developed by ByteDance, aimed at optimizing large model training and inference processes. Versions of VeRL prior to 0.7.0 contained vulnerabilities related to permission licensing and access control. These vulnerabilities stemmed from a...

Adaptive Instruction Composition for Automated LLM Red-Teaming

Many approaches to LLM red-teaming leverage an attacker LLM to discover jailbreaks against a target. Several of them task the attacker with identifying effective strategies through trial and error, resulting in a semantically limited range of successes. Another approach discovers diverse attacks ...

TL-RL-FusionNet: An Adaptive and Efficient Reinforcement Learning-Driven Transfer Learning Framework for Detecting Evolving Ransomware Threats

Modern ransomware exhibits polymorphic and evasive behaviors by frequently modifying execution patterns to evade detection. This dynamic nature disrupts feature spaces and limits the effectiveness of static or predefined models. To address this challenge, we propose TL-RL-FusionNet, a reinforceme...

ARES: Adaptive Red-Teaming and End-To-End Repair of Policy-Reward System

Reinforcement Learning from Human Feedback RLHF is central to aligning Large Language Models LLMs, yet it introduces a critical vulnerability: an imperfect Reward Model RM can become a single point of failure when it fails to penalize unsafe behaviors. While existing red-teaming approaches...

Privacy-Aware Machine Unlearning with SISA for Reinforcement Learning-Based Ransomware Detection

Ransomware detection systems increasingly rely on behavior-based machine learning to address evolving attack strategies. However, emerging privacy compliance, data governance, and responsible AI deployment demand not only accurate detection but also the ability to efficiently remove the influence...

CSLE: A Reinforcement Learning Platform for Autonomous Security Management

Reinforcement learning is a promising approach to autonomous and adaptive security management in networked systems. However, current reinforcement learning solutions for security management are mostly limited to simulation environments and it is unclear how they generalize to operational systems...

Beyond Static Sandboxing: Learned Capability Governance for Autonomous AI Agents

Autonomous AI agents built on open-source runtimes such as OpenClaw expose every available tool to every session by default, regardless of the task. A summarization task receives the same shell execution, subagent spawning, and credential access capabilities as a code deployment task, a 15x...