I am a Senior Research Fellow at Lero, the Irish software research centre, directed by Prof. Lionel C. Briand.

I obtained my Ph.D. from The Chinese University of Hong Kong, Shenzhen in 2025, supervised by Prof. Pinjia He.

My research focuses on trustworthy AI, code agents, and automated testing. I proposed Retromorphic Testing, a technique for automatically constructing test oracles for modern software. My work has appeared at ICML, ICSE, ISSTA, ESEC/FSE, and ACL.

My recent work on UTBoost and SWE-ABS investigates how weaknesses in benchmark tests can inflate the measured success of coding agents. Earlier work develops automated testing methods for image captioning and named entity recognition, and examines the trade-offs between classical and deep learning models for log anomaly detection.

Research and community

Contact

For research discussions and collaboration enquiries:

boxi.yu@lero.ie

You can also paste the address into your preferred email service.

Portrait of Boxi Yu
Boxi YuSenior Research Fellow at Lero, Ireland