Research

My research explores how language models learn, forget, and follow instructions. I’m interested in understanding their behavior and building better ways to evaluate it.

Google Scholar

Publications & preprints

3 papers
2026

WaterDrum: Watermark-based Data-centric Unlearning Metric

Xinyang Lu*, Xinyuan Niu*, Gregory Kang Ruey Lau*, Nhung Bui, Rachael Hwee Ling Sim, John Russell Himawan, Fanyu Wen, Chuan Sheng Foo, See-Kiong Ng, Bryan Kian Hsiang Low

Equal contribution

Using text watermarks to evaluate whether a language model has removed the influence of training data, including when forgotten and retained data are closely related.

Research & collaboration

I’m always happy to discuss research ideas and potential collaborations.

Email me