Blaise Cruz

Low-resource Natural Language Processing

Mabuhay! 👋

I’m a researcher at Samsung Research Philippines where I specialize in NLP problems at the intersection of multilinguality and low-resource languages.

My research interests revolve around creative approaches to solve NLP problems constrained by minimal data. Topics I have worked on encompass a wide variety of tasks such as fake news detection, multilingual translation, language modeling, and linguistic code-switching — all done within the scope of low-resource languages.

Given that my native tongue — Filipino — is a low-resource language, I spend time to develop and open-source datasets and models to boost research in this area. Do check out my Resources tab if you’re interested in pretrained models and benchmark datasets for my language! My Filipino transformer models are also available on 🤗 HuggingFace under my username jcblaise.

Aside from my work at Samsung, I am also a lecturer at the University of the Philippines - Diliman, as well as a research consultant for the DLSU Center for Language Technologies (CELT) where I advise graduate students on dialogue generation research. Prior to this, I also spent time with Senti AI and DLSU COMET working on conversational agents and machine learning.

If you’re interested in collaborating or if you want to chat about low-resource languages, feel free to get in touch! You may reach me through my email me (at) blaisecruz (dot) com.