📋 Job Overview

The role involves serving as a visual domain expert collaborating on innovative native multimodal large model research, including novel model architectures and modeling for '2D+time' and '3D+time' large models. Responsibilities include exploring understanding and generation of the real physical world, multimodal reasoning, and self-evolving continuous learning in large models. The position also entails tracking the latest academic and industry research trends, participating in international conferences and seminars, collaborating with global top teams, and publishing results to the community or transferring them to internal product departments.