Text2TCS's Repositories

Manually corrected Dataset from CogALex VI challange

Scripts and data to recreate mBART termextraction experiments

Extracting terms from text using XLM-R for token and sequence classification

A pipeline approach to automatically extract terminological concept systems from text. We use multilingual neural language models to extract terms and their relations on a an intra-sentence level.

The winning system of the Text2TCS project team submitted to the CogaLex VI shared task.