Large Language Model Security: Model Extraction Attacks Explained
Автор: Fuzzy Labs
Загружено: 2024-02-16
Просмотров: 350
Large Language Model Security: Model Extraction Attacks Explained
Join Matt and Danny as they dive deep into the world of large language model security in a new video in their series. In this episode, they focus on model extraction attacks, a method by which attackers can steal valuable information by querying a model extensively. Understanding the working mechanism of such attacks, the defense strategies against them, and their detection difficulties are also discussed. Watch out for the next episode where 'jailbreaking' will be explained in detail.
00:00 Gangnam Style
00:05 Intro
00:25 What is a model extraction attack?
00:49 How do you steal models?
02:17 How can you defend against it?
03:38 What's next?
03:56 Outtakes
Доступные форматы для скачивания:
Скачать видео mp4
-
Информация по загрузке: