InAIM IntelligencebySangyoon Yu·Aug 30, 2024Automated Multi Turn LLM Jailbreaks — Crescendo from MicrosoftBackground
InAIM IntelligencebySangyoon Yu·Dec 23, 2023Prompt Injection Attack 이해하기Disclaimer: 이 글은 Lakera AI의 prompt injection 블로그 글[링크]을 바탕으로 작성되었습니다.
InAIM IntelligencebySangyoon Yu·Dec 18, 2023Reviewing “Are aligned neural networks adversarially aligned?”In this review, we will look into “Are aligned neural networks adversarially aligned?” published in NeurIPS 2023.
InAIM IntelligencebySangyoon Yu·Dec 10, 2023Reviewing “Automatically Auditing Large Language Models via Discrete Optimization”In this review, I will explain “Automatically Auditing Large Language Models via Discrete Optimization” published in ICML 2023.
InAIM IntelligencebySangyoon Yu·Dec 10, 2023Reviewing “AUTOPROMPT: Eliciting Knowledge from Language Models with Automatically Generated…In this review, I delve into the research paper “AUTOPROMPT: Eliciting Knowledge from Language Models with Automatically Generated…
InAIM IntelligencebySangyoon Yu·Dec 7, 2023Reviewing “On Evaluating Adversarial Robustness of Large Vision-Language Models”In this article, I will review “On Evaluating Adversarial Robustness of Large Vision-Language Models” published in NeruIPS 2023.