The Laboratory of Network Science and Cognitive Intelligence (NSCI Lab) aligns itself with the international frontiers of complex networks and intelligent computing, focusing on the cross-disciplinary integration of "complex graph data, cognitive modeling, and intelligent decision-making" to construct a next-generation graph intelligence research system that harmonizes theoretical depth with practical utility. With complex graph representation learning as its core foundation, the laboratory systematically conducts research on multi-dimensional graph data modeling methodologies, emphasizing three critical dimensions: topological structure, semantic information, and temporal evolution. It explores unified representations and efficient learning methods for multi-scale structural patterns and dynamic behavioral mechanisms within complex systems.

Building upon this, the laboratory further extends its scope to the domain of trustworthy and secure graph modeling. Addressing critical issues such as insufficient robustness, vulnerability to adversarial attacks, and opacity in decision-making of complex graph models in open environments, the lab conducts research on robust generalization mechanisms, security defense methodologies, and trustworthiness enhancement technologies to develop an interpretable, verifiable, and deployable graph intelligence model framework.

Oriented toward national strategic imperatives and the development of the digital economy, the laboratory actively promotes the practical application of graph intelligence in intelligent scenarios, with strategic emphasis on financial security, system security, and social governance. This creates a comprehensive technological chain extending from foundational theoretical breakthroughs to engineering application validation.

In recent years, the laboratory has achieved a series of representative research milestones in complex graph representation learning, dynamic graph modeling, and graph model security. It has proposed multi-level structural modeling methods, heterogeneous semantic fusion mechanisms, and evolution-oriented dynamic graph learning frameworks. These achievements have been validated in practical scenarios such as cyberspace security, internet financial risk control, and public opinion governance.

In the future, the laboratory will continue to revolve around the core axis of "topological understanding – cognitive mechanism modeling – intelligent decision optimization." By driving the deep integration of network science and artificial intelligence, the lab strives to achieve internationally influential and groundbreaking results in foundational complex graph theory and trustworthy intelligent systems.