Elwood Zonghao Ying

Foundation-Model-Paper-Notes.

BAP-Jailbreak-Vision-Language-Models-via-Bi-Modal-Adversarial-Prompt. Jupyter Notebook

Jailbreak_GPT4o. Python

RACE. Python

ClawGuard. ClawGuard is a comprehensive security toolkit designed to mitigate risks associated with autonomous agents, such as OpenClaw and other LLM Agents.

SafeBench. Python

Awesome-Trustworthy-GenAI.

DeepSeek-Safety-Eval. Python

Open-Source-Backdoor-Learning. official/unofficial open source code/dataset for backdoor attack and defense

ATLAS_Challenge_2025. Python

Awesome-Datasets-for-Cybersecurity.

Best_Practice_for_Security_of_AI. Jupyter Notebook

VEIL. Code and data for paper VEIL: Jailbreaking Text-to-Video Models via Visual Exploitation from Implicit Language

AegisLink. 灵盾 — AI Agent 身份与权限中枢

PRISM-Programmatic-Reasoning-with-Image-Sequence-Manipulation-for-LVLM-Jailbreaking. official code

personal-backup. backup

Eval_GPT4o_on_SelectiveTest. Eval Performance of GPT-4o on Selective High School Placement Test

K410ng-s-NG-WAF. The waf is designed by dr0ps，and remould by Yale

Best_Practice_in_AI_for_Security. Best_Practice_in_AI_for_Security

Awesome-material-for-TrustworthyAI.

Evolving-Deception. Evolving Deception: When Agents Evolve, Deception Wins