Expertise

I am currently interested in explainable audio deepfake detection (ADD) with self-evolving audio language models (ALMs).

Nowadays, ALMs are enabled to tackle complex audio understanding and reasoning tasks. The ALM-based methods for ADD should not limited to perform a binary label prediction, but provide the trustworthy acoustic evidences. Moreover, the ADD methods are expected to detect new deepfake audio generated by state-of-the-art audio synthesis methods adaptively.

Therefore, my main focus is to excavate the potential of ALMs for ADD via the self-evolving framework, with the concerns of a trustworthy process for audio reasoning tasks.

Besides, as I have been working on developing evolutionary computation (EC) to solve combinational optimisation problems during my master's time, the EC for discrete optimisation problems also intrigues me.

Organizational Affiliations

School of Computer Science & Electronic Engineering, Faculty of Engineering and Physical Sciences, University of Surrey

Research Student, School of Computer Science & Electronic Engineering, Faculty of Engineering and Physical Sciences, University of Surrey

Education

Computer Science
09/2018–07/2022, Bachelor of Computer Science (BCompSc), China Agricultural University (China, Beijing) - CAU

Supervised by Dr. Lu Jia, AP at China Agricultural University, co-supervised by Prof. Wei-Neng Chen at South China University of Technology.

Computing
09/2022–07/2025, Master of Engineering (MEng, ME or MEng), South China University of Technology (China, Guangzhou) - SCUT

Supervised by Prof. Wei-Neng Chen