Yann LeCun, a recipient of the Turing Award and Meta's Chief AI Scientist, pointed out at the World Economic Forum that generative models are not suitable for video processing, and AI needs to make predictions in abstract spaces. As internet text data becomes scarce, AI researchers are turning their attention to video and realizing the critical importance of understanding causality for future AI systems. Therefore, new models should learn to predict in abstract representation spaces rather than in pixel spaces. The challenge in video processing lies in the complexity of pixel spaces, hence requiring new architectures to handle video inputs and make predictions in abstract representation spaces. To solve the challenges in video processing, it is necessary to create new scientific methods and technologies that enable AI systems to utilize information like humans do.
Yann LeCun: Generative models are not suitable for video processing, AI understanding of video must make predictions in abstract space

站长之家
This article is from AIbase Daily
Welcome to the [AI Daily] column! This is your daily guide to exploring the world of artificial intelligence. Every day, we present you with hot topics in the AI field, focusing on developers, helping you understand technical trends, and learning about innovative AI product applications.