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
#cs.AI
共收录 1652 条相关情报。
← 返回所有主题Harmful SFT Leaves a Continuous Trace in LLM Checkpoint Updates
Safety auditing of post-trained large language models typically relies on model behavior, requiring model execution and
Can Power Draw Constrain Covert Compute? Limits of Analogue Verification for AI Governance
Frontier AI treaties or agreements on limiting computation require external verification; an external auditor must be ab
From Sandbox to Enforcement: Confidence-Qualified Threat Intelligence for Critical Infrastructure
Security operations centres and national incident-response teams defending critical infrastructure collect abundant thre
Polar: LLM-Powered Synthesis of Real-World Cyber Evidence for Prioritization and Mitigation
Cyber threat analysis increasingly depends on evidence distributed across vendor advisories, vulnerability databases, an
SAFESHIELD: A Decision-Organization Framework for Deployment-Time Safety of Small Language Models
Deployment-time safety of language models is commonly implemented through runtime guardrails such as input moderation, r
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
Differentially Private Mixing of Public Datasets Improves Private Learning
Many machine learning applications involve sensitive data and therefore require training under differential privacy (DP)
Does AI Help Cyber Attackers or Defenders? Evidence from Nonpublic Vulnerabilities and Subsequent Attacks
The release decision for frontier AI systems increasingly relies on cyber capability benchmarks, yet public vulnerabilit
GCTAuto-encoder: A Cross modal Framework for Security Flaw Detection in IoT Networks
IoT encompasses diverse physical entities, from smart home devices to autonomous vehicles, creating a complex environmen
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
Towards a Unified Misuse Monitoring Benchmark
LLM agents increasingly act in multi-actor environments, exposing them to misuse from multiple sources: decomposition at
Where Did the Repair First Go Wrong? Localizing the Origins of Silent Failures in Agentic Vulnerability Repair
Localizing where an LLM-based agent first fails to uphold security during a repair can show which stage of its workflow
Cross-Lingual Transferability of Training Data Extraction Attacks to Recover Memorized PII
The robustness of Personally Identifiable Information (PII) protection in Large Language Models (LLMs) is a critical con
Backdooring Sparse Autoencoders
Sparse autoencoders (SAEs) are increasingly used not only to interpret language models but also to intervene on their in
Runaway Reaction: When Benign Skills Compose into Malicious Behavior
Agent skills package task-specific knowledge and procedures that can be composed to support complex agent tasks, while p
Agentic-ZTA: A Multi-Agent Architecture for Autonomous Zero Trust Enforcement
Agentic AI is emerging as a promising paradigm for automating complex cybersecurity decisions, yet its use in enforcing
SimpleMark: Fast Multi-Bit Text Watermarking under f -Divergence Constraints
We introduce a framework for multi-bit text watermarking with security defined directly through $f$-divergence from the
Visual Grounding Safety in Vision-Language Models
Vision-language models (VLMs) are increasingly trained to generate structured outputs like points and bounding boxes tha