AI Research at TechWolf

First and foremost, TechWolf is a brain company. That means we do a lot more than just apply machine learning techniques; rather, we research to advance the state-of-the-art ourselves. This research is done in collaboration with respected institutions like University of Cambridge and Ghent University. With openness being one of the core values of TechWolf, we also aim to contribute to the academic community at large. We do this through publishing a selection of our research, making this work accessible to others while also showing that it meets the standards set by international, peer-reviewed venues. Below, we provide an overview of recent publications.

Van Hautte, Jeroen, Guy Emerson, and Marek Rei. "Bad Form: Comparing Context-Based and Form-Based Few-Shot Learning in Distributional Semantic Models. Proceedings of the 2nd Workshop on Deep Learning Approaches for Low-Resource NLP (DeepLo 2019).

Van Hautte, Jeroen, Vincent Schelstraete, and Mikaël Wornoo. "Leveraging the Inherent Hierarchy of Vacancy Titles for Automated Job Ontology Expansion." Proceedings of the 6th International Workshop on Computational Terminology (2020).

Jens-Joris Decorte, Jeroen Van Hautte, Thomas Demeester, and Chris Develder. "JobBERT: Understanding Job Titles through Skills." International workshop on Fair, Effective And Sustainable Talent management using data science (FEAST) as part of ECML-PKDD 2021.

Jens-Joris Decorte, Jeroen Van Hautte, Johannes Deleu, Chris Develder and Thomas Demeester. "Design of Negative Sampling Strategies for Distantly Supervised Skill Extraction" International workshop on Recommender Systems for Human Resources (RecSys in HR) as part of RecSys 2022.