4 papers
Model-as-a-Trojan: Self-Activating Covert DNS Exfiltration Baked Into the Weights of an LLM Agent
Barak Haryati, Shaked Zychlinski
Open-weight models are downloaded from public hubs and run inside agent harnesses that grant them tools, most powerfully, a shell, under the assumption that a sandbox contains what…
Broken-Token: Filtering Obfuscated Prompts by Counting Characters-Per-Token
Shaked Zychlinski, Yuval Kainan
Large Language Models (LLMs) are susceptible to jailbreak attacks where malicious prompts are disguised using ciphers and character-level encodings to bypass safety guardrails. Whi…
Do Stop Me Now: Detecting Boilerplate Responses with a Single Iteration
Yuval Kainan, Shaked Zychlinski
Large Language Models (LLMs) often expend significant computational resources generating boilerplate responses, such as refusals, simple acknowledgements and casual greetings, whic…
A Whole New World: Creating a Parallel-Poisoned Web Only AI-Agents Can See
Shaked Zychlinski
This paper introduces a novel attack vector that leverages website cloaking techniques to compromise autonomous web-browsing agents powered by Large Language Models (LLMs). As thes…