STAI CDT student Munkhtulga Battogtokh had a successful trip to EuroVis 2024 in May 2024 where he presented the paper ‘Visual Analytics for Fine-grained Text Classification Models and Datasets’ at the main conference. Munkhtulga was also invited to give a talkat the first Visualization for NLP (VIS4NLP) workshop co-located at the conference.
The design study paper ‘Visual Analytics for Fine-grained Text Classification Models and Datasets’. This is a collaboration between me and the co-authors: Yiwen Xing, Cosmin Davidescu, Alfie Abdul-Rahman, Michael Luck, and Rita Borgo. The paper combines explainable AI (XAI) with visual analytics to tackle challenges encountered in applying large language models to real-life problems. It is now published in the journal Computer Graphics Forum (https://doi.org/10.1111/cgf.15098).
The talk at VIS4NLP was titled ‘Explainable AI to Visual Analytics: Putting Explanations to Practice for Fine-Grained Natural Language Understanding’.In this talk, Munkhtulga shared the research journey from explainable AI to visual analytics with experts from the visualization and the natural language processing (NLP) fields.
Munkhtulga was very pleased with how EuroVis 2024 went and he particularly enjoyed the opportunity to explore the location of the conference, Odense, Denmark. He felt a particular connection to the city due to it being the birthplace of Hans Christian Andersen, which was reflected in the names of the hotel rooms where Munkhtulga stayed. As Munkthulga said, ‘I loved the fairy tale (The Ugly Duckling) as a child and I am sure it had an impact on me. As the saying goes, “it takes a village to raise a child”. Odense, as it turns out, is apparently a part of the “village” that raised me’.
Congratulations to Munkhtulga on a successful conference.