TABILAB
Text Analytics and Bioinformatics Lab
We develop machine learning methods for understanding both human language and biological systems.
Welcome to TABILab
TABILab is a research group developing machine learning methods for both language and biological data. Our projects span natural language processing and bioinformatics.
Natural Language Processing (NLP)
We study linguistic structure, information extraction, and language modeling, with a particular focus on Turkish and other low-resource languages.
Bioinformatics
We develop computational representations of proteins, genes, and compounds to study interactions and support molecular discovery. This work includes the Life Language Understanding Lab (LifeLU), connecting natural language processing and computational biology.
Our work is grounded in interdisciplinary collaboration and high-quality datasets that enable robust, theory-informed models across both domains.
Highlights
TURNA
The biggest Turkish encoder-decoder language model up-to-date, available on HuggingFace.
TULAP
Turkish Language Processing Platform - A comprehensive resource for Turkish NLP tools and datasets.
PUFFIN
Protein unit discovery with functional supervision for finding structurally coherent, function-aware units in proteins.