1. Conduct in-depth fundamental and applied research on generative world model technologies, focusing on state-of-the-art advances in video-driven and agent-driven interactive world simulation, long-term temporal consistency modeling, and spatiotemporal causal reasoning.
2. Address fundamental technical bottlenecks in interactive AI content generation, develop robust and high-performance world model research prototypes, and promote the paradigm innovation and industrial deployment of next-generation intelligent simulation systems.
3. Pursue high-impact academic publications, and establish and sustain long-term academic cooperation with top-tier research institutions and scholars worldwide in the fields of world models, video generation, and embodied artificial intelligence.
任职要求
1. Master’s degree or above. Candidates shall possess solid research expertise and practical experience in world modeling, video generation, multi-modal interaction, and spatiotemporal modeling. A proven track record of high-quality peer-reviewed publications in relevant domains is highly preferred.
2. Possess a rigorous mathematical foundation and proficient programming capabilities. Be capable of independently designing, modeling, and analyzing complex algorithms for temporal generation and causal reasoning, with solid proficiency in Python and C++.
3. Demonstrate strong self-motivation and result-driven awareness. Capable of proactively leading technical formulation and tackling core challenges to deliver end-to-end, zero-to-one technical breakthroughs in world model research.
4. Adopt an agile and entrepreneurial working mindset. Able to deliver high-quality research outputs efficiently within a fast-paced, multi-tasking innovative research environment.
