【09-17 PIFI talk】 KLSS Seminar on "Metamorphic Testing Beyond Testing (蜕变测试踰测试)"

Title: Metamorphic Testing Beyond Testing (蜕变测试踰测试)

Speaker: Prof.Tsong Yueh Chen (陈宗岳)

Swinburne University of Technology, Australia

Time: 2026年9月17日（星期四）下午3:00

Venue: 中国科学院软件研究所5号楼 334报告厅

Abstract: Metamorphic Testing (MT, 蜕变测试) was first proposed in late1990’s. In 2021, ISO/IEC/IEEE recognized MT as a standard testing technique – the only new testing technique recognised in the past decade. Today, MT is extensively used and studied. Over hundreds of MT papers were published including many Master and Doctoral theses. A recent survey paper on the quality assurance of AI systems reported MT as the most popular testing method for AI systems.

Though MT was originally proposed as a software testing technique, nowadays it migrated to other domains beyond testing and verification. In this talk, we will illustrate how MT has evolved to become a useful method in many areas other than testing. We will discuss how to train deep neural networks in a systematic way using MT, how to extend the concept of program slicing through its integration with MT, how to extend the applicability scope of existing debugging and program repairing methods, how to complement the conventional coverage metrics, how MT has evolved into an assessment and understanding method, in particular to the AI systems

We will also provide new insights and potential research directions for MT.

Bio: Tsong Yueh Chen (陈宗岳) is currently a Professor of Software Engineering at Swinburne University of Technology, Australia. Prior to joining Swinburne, he taught at The University of Hong Kong and The University of Melbourne.

Chen received education at The University of Hong Kong, Imperial College London and The University of Melbourne.

Chen is widely recognized as the Father of Adaptive Random Testing as well as the Father of Metamorphic Testing. He is the 28th recipient of the ACM SIGSOFT Outstanding Research Award, and is the first Chinese to receive this award. His award citation is “for contributions to software testing through the invention and development of metamorphic testing”. He is an IEEE Fellow with citation “for contributions to software testing through the invention of metamorphic testing and adaptive random testing”.

In 2026, Chen was inducted to the inaugural class of the ACM SIGSOFT Software Engineering Academy that was established as part of the Fifty Years of ACM SIGSOFT celebrations.

His main research interest is software testing. In addition to inventing adaptive random testing and metamorphic testing, Chen co-invented with Y. T. Yu the software testing technique of Proportional Sampling Strategy to address the problem on how to make partition testing better than random testing. With H. Leung, they invented the technique of Generalized Proportional Sampling Strategy. With R. Merkel, they proved an important result on the bound of testing effectiveness. With T. H, Tse, F.-C. Kuo and R. Merkel, they proposed and developed a new software testing area of failure-based testing that includes Proportional Sampling Strategy and Adaptive Random Testing.