У нас вы можете посмотреть бесплатно AI to Automatically Fix Security Bugs by Google или скачать в максимальном доступном качестве, которое было загружено на ютуб. Для скачивания выберите вариант из формы ниже:
Если кнопки скачивания не
загрузились
НАЖМИТЕ ЗДЕСЬ или обновите страницу
Если возникают проблемы со скачиванием, пожалуйста напишите в поддержку по адресу внизу
страницы.
Спасибо за использование сервиса savevideohd.ru
Jan Nowakowski, a Machine Learning Engineer at Google, delivered a talk for the @BuzzRobot community about how he and his colleagues developed a pipeline for automated bug fixes. Their work specifically targets security related bugs. They fine-tuned a large language model (LLM) based on Gemini Pro, which helped the engineers successfully fix around 15% of bugs. You may think it's a low success rate but it's definitely much higher than the team expected when they started working on the project. Timestamps: 0:00 Introduction. The lecture covers security related bug fixes. 6:52 Description of the AI-powered pipeline to automate vulnerability fixing and explanation of each of these steps. 20:15 The results of the LLM performance. The fix rate of the LLM. 22:10 Overview of other projects that use LLMs for bug fixes. 25:33 The Q&A. Social Links: Newsletter: https://buzzrobot.substack.com/ X: https://x.com/sopharicks Slack: https://buzzrobot.slack.com/join/shar...