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
#cs.AI
共收录 1652 条相关情报。
← 返回所有主题From Reactive Containment to Proactive Assurance: Lessons from OpenAI, Anthropic, and Google Agent Security Incidents
In 2026, cybersecurity evaluations involving OpenAI, Anthropic, and Google agents reached real systems outside their aut
ReSI: Recursive Safety Improvement toward Resistant and Resilient AI
Recursive self-improvement, the participation of AI systems in improving their own capabilities, is beginning to move fr
LTBD: Learnable Trust-Boundary Delimiters for Prompt Injection Defense
Large language models (LLMs) perform remarkably well on complex tasks, yet remain highly vulnerable to prompt injection
GROB: A Multi-Agent Architecture for Public-Trace Investigation of Candidate Agentic Activity
We present GROB, a multi-agent architecture for investigating candidate autonomous-agent activity through public Interne
Characterizing Statistical Separability in TP-CRIV for Probabilistic AI Models
Third-party challenge-response identity verification (TP-CRIV) enables an independent verifier to assess whether a claim
Language Modeling is Monotone Compression
A long-standing hypothesis in artificial intelligence and neuroscience posits that intelligence is closely related to co
NOMOS: Compiling Written Policies into Statically Verified Tool-Call Gates for LLM Agents
Tool-using LLM agents violate the policies they are deployed to enforce, often silently. Prior defenses hand-write rules
Speedbumps: Rejection Attacks on Speculative Decoding
Speculative decoding is a popular technique for increasing the speed and reducing the costs of large language model (LLM
BRANCH: Bypassing Multi-Scanner AI Guardrails
AI systems increasingly rely on Large Language Models (LLMs) as core reasoning engines, making them targets for prompt i
Nullify: Null-Space Activation Steering for Training-Free LLM Unlearning
Large Language Models (LLMs) inevitably internalize substantial amounts of sensitive or private information during pre-t
SLDR: Defending Against Malicious Fine-tuning via Selective Layers Recovery and Dynamic Routing
Fine-tuning-as-a-service enables users to adapt aligned large language models (LLMs) to specialized tasks, but malicious
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
Defensive Sufficiency in a Stackelberg Model of AI Security
Feedback from automated testing, human red teaming, and incident response can strengthen an AI system's defenses when di
When AI Finds Hidden Messages, Does It Report?
When an assistant encounters a message for another AI, does it tell its user? Four fixed model-provider deployments perf
MRCert: Towards Post-deployment Patch Robustness Certification for Adversarially Patched Samples via Type-specific Masking
In post-deployment time, inputs to deep learning models may or may not be adversarially patched. Patch robustness certif
Correct Answers, Unsupported Findings: Evidence Binding in Forensic Reconstruction of LLM Agent Logs
Forensic reconstruction of LLM-agent actions requires not only recovering the correct value, but establishing which pres
MARS: Malware Analysis with Rule-Based Scoring of LLM Claims
Large language models can triage malware through direct verdicts or behavioral claims scored by an external policy. We p
Secure-CUA: Controlling Untrusted Influence in Computer-Use Agents
Computer-use agents (CUAs) perform tasks across applications (such as desktops, mobile apps, and web browsers) by observ
From Investigation Failures to Reliable SOC Agents: Understanding and Improving LLM-Based Alert Triage
Security operations centers (SOCs) must triage large volumes of alerts, most of which are benign, while missed attacks c