ATTCK-R: Knowledge-Aware Long-Horizon Jailbreaking for Protected LLMs
As large language models are deployed behind increasingly sophisticated safety mechanisms, evaluating their robustness u
#cs.CL
共收录 424 条相关情报。
← 返回所有主题When Agents Talk, Secrets Travel: Topology-Driven Leakage and Recipient-Conditioned Sanitization in Multi-Agent LLM Systems
Multi-agent LLM systems (MALS) introduce a privacy-leakage problem that conventional sensitivity-based defenses miss, si
Exploring Jailbreak Vulnerability Across Depth-Based Dynamic Inference in Large Language Models
Large Language Models (LLMs) have been shown to be computationally expensive and vulnerable to jailbreak attacks, in whi
APEX-Harm: Evaluating Agent Safety in Shared Professional Workspaces
Professional agents increasingly work through command-line interfaces and multiple tools over shared workspaces containi
VEPA: Verified Evolving Formal Policy for Lifelong LLM Agent Guardrails
LLM agents increasingly operate over sensitive resources, such as private documents and mutable system states, which can
Cancerous Backdoor Attack: Gradual Activation via Invisible Tokens and A Stage Controller
Backdoor Attacks on Large Language Models (LLMs) typically follow a trigger-response pattern, where satisfying a trigger
Adversarial Cues in Decision Models Used as Judges: The Role of Request Presentation
An answer judge instructed to grade the final commitment should reject an explicitly wrong final value even when an earl
PatchBench: Measuring Collateral Damage in Activation Patching
An LLM safety patch can pass a benchmark while still being a poor repair. This risk is especially acute for jailbreak re
Sensitive-Topic Leakage Through LLM Routing Metadata: Measurement and Mitigation
LLM routers pick a cheap or expensive model per request by its content, and many gateways and some cloud platforms can l
Constrained-Action AI Remediation for SIEM/XDR via a NeMo-Guardrails Proxy
Security Operations Centers (SOCs) for information technology and operational technology share one incident-response pro
Constitution-Guided Watermarking
Watermarking enables language model providers to identify text generated by their models. However, its desired propertie
Adversarial Images Hijack Web Agents from Visual Grounding to Browser Execution
Modern web agents built on large vision-language models process webpages, select relevant UI elements, and translate mod
Quad-State Safety Evaluation of Open-Weight Large Language Models on Non-Canonical Inputs
Standard safety evaluations of large language models assess harmful requests written in canonical plain text, while mode
How Fragile Is On-Device Language Model Safety? Localizing Safety-Critical Parameters for Sparse Fault Analysis
As small language models (SLMs) are increasingly deployed on resource-constrained and on-device platforms, including as
Where Rules End and Judges Begin: Measuring the Judgment Boundary in Multi-Agent Systems Security
LLM-based multi-agent systems (MAS) engage tools, share memory, and delegate tasks, often encountering adversarial conte
Safeguarding LLMs via Model-Agnostic Latent Safety Signals from Dark Knowledge
LLMs have advanced rapidly, raising growing concerns about their safety. Recent work has proposed approaches to detect a
Not What a Child Expressed: Auditing the Sign-to-Text Safety Interface in Child-Facing AI
Automatic sign language translation (SLT) has entered consumer products, turning American Sign Language into English tex
Lineage-Aware Memory Governance: A Derivation-Gated Framework for Privacy-Preserving Column-Level Access Control in Enterprise AI Agents
Enterprise AI agents that share a memory store face two unaddressed risks: sensitive data can leak through legitimately
Jailbreaking Open-Weight LLMs via Random Embedding Perturbations
While open-weight models have enjoyed steady progress in capabilities and wide adoption across multiple domains, their s
RAISED: Self-Distillation for Robustness to Prompt Injection in LLM Agents
Tool-using language-model agents are vulnerable to indirect prompt injection because they must act on untrusted external