Greetings World!

I am a team lead at DSO National Laboratories pretraining multilingual large language models, contributing across data, training, evaluation and finetuning, and building agentic retrieval-augmented products. We continued pretrained a multilingual large language model from LLaMa-2 7B that’s competitive with Sailor-7B from SEA AI Lab on both english and multilingual benchmarks.

I graduated from National University of Singapore Computer Science. I was also part of the Turing Research Programme.

Previously, I researched on semi-supervised text classification at DSO under Hai Leong and Prof Wee Sun. I also researched on Bayesian Optimisation at NUS under A/P Bryan Low.

News

310522: I’ve graduated!

160122: I’ve submitted a short paper for review at NAACL 2022 🥳! I owe much to my supervisors Dr Chieu and Prof Wee Sun.

170521: I’ve published my automatic github issue labeller in the market place. 😄