会员服务
KVEraser: Learning to Steer KV Cache for Efficient Localized Context Erasing
Arxiv
0+阅读 · 6月15日
Defending against Adaptive Prompt Injection Attacks via Reasoning-enabled Task Alignment
Arxiv
0+阅读 · 6月13日
Send a SCOUT First: Pre-hoc Reasoning for Adaptive Detector Allocation in Prompt-Injection Defense
Arxiv
0+阅读 · 6月14日
From ASR to ASP: Evaluating Prompt Attack Vulnerabilities Against Open-Source LLMs
Arxiv
0+阅读 · 6月14日
Cordyceps: Covert Control Attacks on LLMs via Data Poisoning
Arxiv
0+阅读 · 6月15日
MUZZLE: Adaptive Agentic Red-Teaming of Web Agents Against Indirect Prompt Injection Attacks
Arxiv
0+阅读 · 6月14日
AttriGuard: Defeating Indirect Prompt Injection in LLM Agents via Causal Attribution of Tool Invocations
Arxiv
0+阅读 · 6月10日
It's a TRAP! Task-Redirecting Agent Persuasion Benchmark for Web Agents
Arxiv
0+阅读 · 6月4日
Device Context Protocol: A Compact, Safety-First Architecture for LLM-Driven Control of Constrained Devices
Arxiv
0+阅读 · 5月24日
No Hidden Prompts Needed! You Can Game AI Peer Review with Presentation-Only Revisions
Arxiv
0+阅读 · 6月11日
MalSkillBench: A Runtime-Verified Benchmark of Malicious Agent Skills
Arxiv
0+阅读 · 6月9日
Can It Reach the Generator? Investigating the Survival of Prompt-Injection Attacks in Realistic RAG Settings
Arxiv
0+阅读 · 5月28日
Assessing Automated Prompt Injection Attacks in Agentic Environments
Arxiv
0+阅读 · 6月9日
Investigating Detection and Obfuscation of Prompt Injection Attacks Against Software Reverse Engineering AI Agents
Arxiv
0+阅读 · 5月29日
GitInject: Real-World Prompt Injection Attacks in AI-Powered CI/CD Pipelines
Arxiv
0+阅读 · 6月7日
参考链接
微信扫码咨询专知VIP会员
Top