The Computational Linguistics and Text Mining Lab (CLTL) is lead by Prof. Dr. Piek Vossen and Prof. Dr. Antske Fokkens at the Faculty of Humanities of Vrije Universiteit Amsterdam. The CLTL has evolved from conducting research on Computational Lexicology and Terminology to dealing more broadly with AI and Computational Linguistics.

Our key research themes are:

• Identity, reference and perspectives

  The CLTL has a strong track record in conversational AI, but we are also developing Dutch FrameNet, and work on entity and event detection, hate speech and subjectivity.

• Meaning representation

  The CLTL is a long-standing participant of the Global Wordnet Association and a contributor to the Global Wordnet Grid for Dutch. We also study meaning in distributional representations, contextual embeddings and multilingual representations.

• LLMs and interpretability

  The CLTL actively studies, builds and uses Large Language Models. We notably research hallucination and bias in these models and how to correct them.

Engagement and partnership

The CLTL actively engages in interdisciplinary projects with other academic institutions and companies. Medical NLP and Digital Humanities are recurring themes in these partnerships.

The CLTL offers two master programs in AI and NLP, a one-year Text Mining MA, and a two-year Human Language Technology Research MA. The CLTL is a partner in the Creative Technologies Bsc together with the University of Twente. Our teaching staff further give other BA and MA courses at Vrije Universiteit Amsterdam.

We value open science and publish our models, software and resources open source.