Typological Diversity in NLP: What, Why and a Way Forward

To justify the generalisability of multilingual NLP research, multilingual language technology is frequently evaluated on ‘typologically diverse’ language selections. Yet, what this means often remains vague. In this talk, I first discuss what typological diversity means in NLP, and why it matters. Then, I introduce a framework for systematic language sampling, which is inspired by typological insights. Ultimately, the aim of this talk is to inspire research on linguistic typology in NLP that goes beyond merely leveraging databases, but rather incorporates research methodologies from linguistic typology.