An Evaluation of Data Leakage Risks in Tool-Using LLM Agents in Realistic Scenarios
Explorar
Noticias de IA
1359 elementos — filtrados, clasificados y sin duplicados
TrustErase: Auditable Instant Machine Unlearning with Passport-Embedded Representations
Security and Privacy Prompts in the Wild: What Users Ask LLMs and How LLMs Respond
Breaking the Code: Security Assessment of AI Code Agents Through Systematic Jailbreaking …
Adversarial Attacks Leverage Interference Between Features in Superposition
BadScientist: Can a Research Agent Write Convincing but Unsound Papers that Fool LLM Revi…
Membership Inference Attacks against Large Audio Language Models
Like a Hammer, It Can Build, It Can Break: Large Language Model Uses, Perceptions, and Ad…
Learning Red Agent Policy from Observations for Neurosymbolic Autonomous Cyber Agents
Timestamp-Aware Spatio-Temporal Graph Contrastive Learning for Network Intrusion Detection
Graph neural networks at war: integrating cybersecurity and drone intelligence in the Isr…
SkillJect: Effectively Automating Skill-Based Prompt Injection for Skill-Enabled Agents
An AI Security Agent for Banking: Multi-Vector Fraud and AML Detection Across Retail and …
This Copilot vulnerability could expose emails, 2FA codes, and other sensitive data
Securing the future of AI agents
Critical Copilot vulnerability allowed hackers to seal 2FA code from users
AI Red Teaming Explained: What It Is and Why You Need It
Dual-Granularity Orthogonal Disentanglement for Generalizable Audio Deepfake Detection
The Proxy Knows Too Much: Sealing LLM API Routers with Attested TEEs
Communication-Efficient Verifiable Attention for LLM Inference
SkillVetBench: LLM-as-Judge for Multi-Dimensional Security Risk Evaluation in Open-Source…
AgentLeak: A Benchmark for Internal-Channel Privacy Leakage in Multi-Agent LLM Systems
Computational Safety for Generative AI: A Hypothesis Testing Perspective
Robust Spoofed Speech Detection via Temporal Pyramid Modeling
Are Neuro-Inspired Multi-Modal Vision-Language Models Resilient to Membership Inference P…
Can We Stop Malicious AI? KILLBENCH: A Benchmark for External AI Kill Switch Feasibility
A Survey on Agentic Security: Applications, Threats and Defenses
Discrete optimal transport is a strong audio adversarial attack
Automated jailbreak attack targeting multiple defense strategies
GAS-Leak-LLM: Genetic Algorithm-Based Suffix Optimization for Black-Box LLM Jailbreaking