Agentic intelligence is now a new paradigm and a core component of the frontier in artificial intelligence research. Agentic intelligence transforms the capabilities of humans to agents in an automatic way. Unlike earlier systems that responded to a single instruction prompt and stopped to wait for further orders from users, modern agentic systems generally pursue a final target over many steps across multiple turns in the long context: they plan, call functions and tools, write and execute code, search for information, evaluate their own intermediate results, and revise their approach as a comprehensive workflow.
Such a new trend moves AI from a passive “assistant” toward an active collaborator that can carry out entire workflows with limited human supervision, compressing multiple tasks that once took weeks into hours. Yet, the same autonomy remains to be a double-edge sword, which makes agents powerful also raises new challenges in the meanwhile. Errors can accumulate across long chains of actions, agents may pursue objectives in unexpected ways, and responsibility for their decisions becomes harder to determine and align with real world protocols. In this talk, we will introduce our recent studies on agentic intelligence empowered by data and workflow automation as well as some of the challenges and outcomes.

Rui Yan is a full professor at Wuhan University. He was also recognized as a StarTrack scholar in Microsoft Research Asia and Beijing Academy of Artificial Intelligence (BAAI) scholar. He has published more than 100 research articles with more than 23,000 citations to date. His research interests include Natural Language Processing, Large Language Model and Agents. He regularly serves as a senior area chair/senior PC for top-tier venues and has been invited to give tutorial talks in a series of well-known conferences.