10/21/2026 Rudy San Miguel
Written by Rudy San Miguel
Bowen Jin (CS Ph.D. ’26) received the 2026 ACM SIGKDD Doctoral Dissertation Award for his Ph.D. thesis, “Information-seeking Intelligence with Large Language Models: Retrieval, Reasoning and Agentic Learning.” Jin is presently a research scientist at OpenAI where his interests include reinforcement learning, reasoning and AI agents. Jin was advised by Jiawei Han, CS professor and Michael Aiken Chair, while at the Siebel School of Computing and Data Science in the Grainger College of Engineering.
The ACM SIGKDD Dissertation Award recognizes outstanding work done by graduate students in the areas of data science, machine learning and data mining. SIGKDD’s mission is “to provide a forum for advancement, education and adoption of the ‘science’ of knowledge discovery and data mining from all types of data stored in computers and networks of computers.”
Jin’s thesis centers on how to build AI systems that actively seek out information and use it to solve complex problems.
“A useful research assistant needs to find relevant evidence, understand how different pieces fit together and decide what to investigate next,” Jin said. “My research develops methods that help language models do these things more effectively.”
Jin’s work brings together several parts of that problem: helping models understand text and the relationships connecting it, improving how they find and reason over information and teaching them to use search tools over multiple steps. Jin says the connections between scientific papers can be as informative as the words in an individual paper.
“Finding a useful document is only part of the challenge,” Jin said. “A model must also avoid being distracted by other results that look relevant but do not answer the question.”
Jin’s central project, Search-R1, was published in 2025. His former advisor, Han, pointed out that Search-R1 has attracted over 1600 citations and is followed by industry and academia internationally.
“His overall research has received over 5000 citations,” Han said. “Bowen's research is creative and impactful.”
Search-R1 trains language models to alternate between reasoning and searching. Through reinforcement learning, the model learns from feedback on whether it reaches the correct answer, without needing a person to label every intermediate step.
“The broader goal is to make AI a more capable research assistant that can gather evidence and work through questions requiring several rounds of investigation,” Jin said.
Since much of Jin’s development as a researcher happened in the data mining community, he says this award is particularly meaningful to him.
“It encourages me to keep pursuing the questions that connected my Ph.D. projects,” Jin said. “How do we help people find useful knowledge in large amounts of information, and how can AI take a more active role in that process?”
Jin added that this recognition is a shared effort, pointing to his lab mates who helped shape the ideas for his thesis, as well as his collaborators, family, partner and, of course, Han.
“Professor Han had an enormous influence on that process,” Jin said. “He encouraged me to choose problems that could grow into a sustained line of research, and he consistently emphasized ‘real problems, real data and real solutions.’”
Jin added that Han’s advice helped him see connections across his projects and develop a broader direction that became his dissertation. Han’s support extended beyond the research, “from learning to write and present papers to navigating important career decisions.”
Jin also cites the Data and Intelligent Systems (DAIS) lab at Siebel School as key to his development. He says Yu Meng (CS Ph.D. ’23), assistant professor at University of Virginia, and Yu Zhang (CS Ph.D. ’24), assistant professor at Texas A&M University, guided him through his earliest projects and helped him understand how to conduct meaningful research. Meng won the 2024 ACM SIGKDD Doctoral Dissertation Award; Zhang earned the 2025 Runner Up. Meng and Zhang were also advised by Han.
“Looking back, the relationships I built in Urbana-Champaign are as important to me as the papers,” Jin said. “I left Illinois with greater confidence as a researcher and a deep appreciation for the people who helped me grow.”
Jin’s own body of research, along with his collaborations and connections throughout his career, have instilled in him the importance of sharing research openly. His Search-R1 is an example of that work ethic. Jin and his team released code, models and data so that others could study and build on their work.
“Seeing the project support follow-up research has been especially rewarding,” Jin said. “It gives an individual project a life beyond the original paper and allows a much broader community to explore the questions it raises.”
Jin looks forward to his continued role in AI research that began with his work at Siebel School.
“I am excited about AI systems that can help people investigate difficult questions and make new discoveries,” he said. “The goal of making those systems more capable and reliable continues the direction I began exploring at Illinois.”
Grainger affiliations:
Jiawei Han is an Illinois Grainger professor of computer science and a Michael Aiken Chair professor in the Siebel School of Computing and Data Science.