Black-box Membership Inference Attacks on the Pre-training Data of Image-generation Models
Explorar
Noticias de IA
1359 elementos — filtrados, clasificados y sin duplicados
Cordyceps: Covert Control Attacks on LLMs via Data Poisoning
Lessons from Penetration Tests on Large-Scale Agent Systems
Jailbreak susceptibility prediction and mitigation via the behavioral geometry of models
Tracing the Dynamics of Refusal: Exploiting Latent Refusal Trajectories for Robust Jailbr…
SkillSieve: A Hierarchical Triage Framework for Detecting Malicious AI Agent Skills
Turning Bias into Bugs: Bandit-Guided Style Manipulation Attacks on LLM Judges
Furina: Fragmented Uncertainty-Driven Refusal Instability Attack
MRMMIA: Membership Inference Attacks on Memory in Chat Agents
Disentangling Adversarial Prompts: A Semantic-Graph Defense for Robust LLM Security
Millions of AI agents imperiled by critical vulnerability in open source package
FBI agent explains how easy it is to ID people posting AI porn without consent
Do Modern Post-Hoc Watermarking Methods Beat Broken-Arrows?
Black-box Membership Inference Attacks on the Pre-training Data of Image-generation Models
BNP Paribas Works With Mistral to Prep for Mythos-Like AI Models
Committed SAE-Feature Traces for Audited-Session Substitution Detection in Hosted LLMs
Safety in Embodied AI: A Survey of Risks, Attacks, and Defenses
Security of OpenClaw Agents: Fundamentals, Attacks, and Countermeasures
When the Manual Lies: A Realistic Benchmark to Evaluate MCP Poisoning Attacks for LLM Age…
Attested Tool-Server Admission: A Security Extension to the Model Context Protocol
Enhancing Reliability in LLM-Based Secure Code Generation
SoK: A Comprehensive Security Analysis of Jailbreak Resilience in GPT and DeepSeek Models
Explainable Attention-Guided Stacked Graph Neural Networks for Malware Detection
Membership Inference Attacks on Tokenizers of Large Language Models
LivePI: More Realistic Benchmarking of Agents Against Indirect Prompt Injection
Hidden-State Privacy Has an Empty Middle
Jailbreak to Protect: Buffering and Reinforcing via Temporary Jailbreaking for Safe Fine-…
Uncovering Vulnerabilities of LLM-Assisted Cyber Threat Intelligence
Concept Drift Adaptation Using Self-Supervised and Reinforcement Learning In Android Malw…
Batch Normalization Amplifies Memorization and Privacy Risks