Towards Robustness of Text-to-Visualization Translation against Lexical and Phrasal Variability

arXiv:2404.07135v1 Announce Type: new
Abstract: Text-to-Vis is an emerging task in the natural language processing (NLP) area that aims to automatically generate data visualizations from natural language questions (NLQs). Despite their progress, existing text-to-vis models often heavily rely on lexical matching between words in the questions and tokens in data schemas. This overreliance on lexical matching may lead to a diminished level of model robustness against input variations. In this study, we thoroughly examine the robustness of current text-to-vis models, …

abstract arxiv cs.ai cs.cl data data visualizations generate language language processing natural natural language natural language processing nlp processing progress questions robustness text tokens translation type visualization words