Kairos: Transactional Security Enforcement at the System Boundary for LLM Agents
Large language model (LLM) agents are becoming an integral part of the software development lifecycle, and the capabilit
2026-10-14 归档
今日共收录 1875 条安全情报，包含 867 个 CVE，609 篇安全通告，以及 100 篇研究论文。
← 返回归档列表FedGraph: Defending Federated Large Language Model Fine-Tuning Against Backdoor Attacks via Graph-Based Aggregation
Federated fine-tuning of large language models (LLMs) allows multiple clients to collaboratively adapt models without sh
Jev-Guard: Fast Routing and Auditing Agent Actions over Editable Policy Trees
Agent harnesses require a low-latency guardrail model together with customizable audit policies. However, deploying an a
Encrypted Skills Can Become New Security Exploits If AI Agents Know How to Decipher
Third-party skills can expose LLM agents to malicious instructions hidden by encoding. We introduce Ciphered Skill Injec
CompanionSafe: From Multi-turn Safety Evaluation to Guardrails for AI Companions
AI companions provide users with emotional support and ongoing companionship, but their sustained relational interaction
Cost-Efficient Privacy Enforcement for Email-Generating LLM Agents under Strategic Information Requests
LLM agents now act as delegated communication intermediaries: they read private context, draft messages, and send extern
ProtoArena: Can AI Agents Discover and Exploit Real-World Cross-Component Protocol Vulnerabilities?
Large language model (LLM) agents are increasingly used for vulnerability analysis, yet existing benchmarks provide limi
Same Attack, Different Judge: A First Controlled Adversarial Comparison of Jev, LLM Judges, and Purpose-Built Safety Classifiers
TypeSafe AI's Jev is a new kind of safety judge. It generates no text and no label: it answers questions the developer d
SkillScreen: High-Recall Screening of Malicious Agent Skills with Decision Models
Agent skills turn natural-language instructions into actions, creating a security challenge: legitimate automation and m
COMPASS: Membership Inference Attack Against Vision Large Language Models via Adaptive Noise Calibration
Vision large language models (VLLMs) are deployed in privacy-sensitive domains, raising training-data exposure concerns.
TRAP: Tree-Guided Reinforcement-Augmented Probing for Automated LLM Jailbreaking
Aligned large language models (LLMs) remain vulnerable to jailbreak prompts that circumvent their safety guardrails, mak
StableJudge: A Lightweight Adversarial Defense Framework for LLM-as-a-Judge
Large language models (LLMs) are increasingly used as judges to assign scalar scores that drive downstream decision-maki
Logit Leakage through Floating-Point Bias Addition
Language model APIs can expose token-specific biases while withholding logits and token probabilities. Existing bias-bas
CircuitBreak: Pruning a Sparse Safety Circuit of Neurons
White-box neuron pruning has shown that refusal in large language models (LLMs) can depend on a small, removable set of
Purpose IR: Compiling User Purpose for Privacy-Preserving Agent Workflow State
LLM-agent workflows routinely transform rich observations into plans, summaries, memories, handoffs, and other reusable
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
DelegateGuard: Purpose-Bound Capabilities for Delegated LLM Agent Workflows
\textbf{(Context)} LLM agents increasingly delegate subtasks while holding authority over sensitive data, creating confu
AutoSec-RAN: Specification-Grounded LLM Fine-Tuning for Automated Security Analysis of O-RAN
This paper presents AutoSec-RAN, a specification-grounded large language model (LLM) framework for automated security tr
Scalable Weak-to-Strong Verifier Hardening for Agent Environments
Agent environments are essential for large language model (LLM) development. They use programmatic verifiers to turn com
What the Model Cannot See: Covert and Provable Text Invisibility in LLMs
Text wrapped in a trigger can be made behaviorally invisible to a language model: omitted from summaries and quotations,