The DisAI project recently organized its second scientific webinar on the topic of Multilingual NLP: Overview and Low-Resource Multilingual Processing. The webinar was led by Dr. Simon Ostermann (Senior Researcher at DFKI) and Cristina España-Bonet (Senior Researcher at DFKI).
The webinar on Multilingual NLP hosted by the German Research Center for AI (DFKI) was part of the webinar series in the context of the EU project DisAI. The webinar consisted of two parts.
In the first part, we gave an overview of common approaches and techniques in multilingual and cross-lingual natural language processing, with a specific focus on multilingual representations and encodings, as well as non-societal biases in multilingual models.
The second part focused on low-resource multilingual processing and covered self-supervised machine translation as an example of data-efficient processing, as well as adapters and prompt tuning as examples for parameter-efficient techniques, both of which are important building blocks of low-resource NLP.
The two talks presented basic approaches on the two topics, but also gave insights into current research conducted at the Multilinguality and Language Technology Lab at DFKI. Participants had the chance to discuss questions with the lecturers in two discussion blocks.
You can watch the webinar recording here: