T. Zhang Receives Amazon Research Award
09-14-2026
Tianyi Zhang, assistant professor of computer science at Purdue University
Tianyi Zhang, assistant professor of computer science at Purdue University, has received a 2026 Amazon Research Award to support his work on scaling interprocedural data-flow analysis using large language models (LLMs).
Zhang’s project focuses on leveraging LLMs to enhance interprocedural data-flow analysis, a technique used to track how information moves across functions and modules in a program. The technique can help detect issues such as data leaks and unsafe memory operations by identifying where data originates and where it flows.
Modern software systems are increasingly large and complex, often involving dynamic behavior and extensive use of external libraries. Traditional analysis techniques cannot feasibly explore every possible execution path or maintain full context throughout a large program, which can result in missed bugs and false alarms.
“Our work uses large language models to help fill in those gaps, such as prioritizing important paths instead of exhausting all paths and inferring behavior of external libraries based on their documentation,” Zhang said. “This allows us to scale this technique to real-world, industrial codebases while improving both accuracy and coverage.”
Rather than replacing traditional program analysis, Zhang’s approach combines its precision with the ability of LLMs to interpret code and recognize patterns and intent.
“Traditional program analysis relies on heuristics and formal reasoning, which are precise but struggle with ambiguity, like dynamic function calls or missing source code,” Zhang said. “LLMs bring a complementary strength: They can interpret code in a more flexible, human-like way by recognizing patterns and intent.”
The approach could benefit large-scale software systems, including cloud services, mobile applications, operating systems and embedded software. It could help developers detect security vulnerabilities, improve code quality and better understand legacy or third-party components.
“For organizations like Amazon, which maintain massive and diverse codebases, this technology can speed up bug detection, reduce false alarms, and help developers better understand how their systems behave,” Zhang said.
The Amazon Research Awards program recognizes faculty conducting innovative research with potential real-world impact and provides unrestricted gifts to support projects. These awards foster collaboration between academic researchers and Amazon scientists while enabling research in computer science.
About the Department of Computer Science at Purdue University
Founded in 1962, the Department of Computer Science was created to be an innovative base of knowledge in the emerging field of computing as the first degree-awarding program in the United States. The department continues to advance the computer science industry through research. U.S. News & World Report ranks the department No. 16 and No. 15 overall in undergraduate and graduate computer science, respectively. Graduates of the program are able to solve complex and challenging problems in many fields. Our consistent success in an ever-changing landscape is reflected in the record undergraduate enrollment, increased faculty hiring, innovative research projects, and the creation of new academic programs. Learn more at cs.purdue.edu.