Title: Consistency is All We Need

Report time: May 9th, 2025 (Friday) 9:00-10:00

Report location: College Atrium​, DISEI

Inviter: Jun SUN

Abstract: 

Defining "correctness" has long been a central concern in the application of formal methods. The emergence of large language models (lLMs) further complicates traditional notions of correctness, particularly in domains where formal definitions are inherently unattainable, In this talk, l will argue that in many contexts-especially those involving LLMs-consistency, rather than correctness, should serve as the primary standard. Through two case studies-one involving the application of LLMs to a software engineering task and another addressing an Al safety challenge-l will demonstrate how prioritizing consistency enables practical and effective solutions in settings where correctness is difficult or impossible to define.

Biography:

Professor at the Singapore Management University (SMU), Sun Jun earned his Bachelor’s and Doctoral degrees in Computer Science from the National University of Singapore (NUS) in 2002 and 2006, respectively. He has been a faculty member since 2010. His research focuses on AI security, software engineering, formal methods, and the design of algorithms to address challenging real-world problems. Adept at bridging theoretical innovation with practical applications, he demonstrates a strong commitment to advancing both academic research and societal well-being.

Sun Jun has received numerous accolades, including two Lee Kuan Yew Fellowships, recognition as one of the most influential contributors to the field. His scholarly output includes over 100 peer-reviewed papers published in CCF-A category journals and conferences, complemented by multiple ACM Distinguished Paper Awards. Notably, the PAT model checker he co-developed was honored with the “ICFEM 20-Year Most Influential System Award” for its transformative impact on formal verification methodologies.

For further details, please visit: https://sunjun.site