TaTa: A Multilingual Table-to-Text Dataset for African Languages
This addresses the problem of data scarcity for multilingual natural language generation, particularly for African languages, though it is incremental as it extends existing dataset creation methods to new languages.
The authors tackled the lack of multilingual table-to-text datasets by creating TaTa, a large dataset focusing on African languages, which includes 8,700 examples in nine languages and is shown to be challenging for current models, with less than half of outputs from an mT5-XXL model being understandable and attributable.
Existing data-to-text generation datasets are mostly limited to English. To address this lack of data, we create Table-to-Text in African languages (TaTa), the first large multilingual table-to-text dataset with a focus on African languages. We created TaTa by transcribing figures and accompanying text in bilingual reports by the Demographic and Health Surveys Program, followed by professional translation to make the dataset fully parallel. TaTa includes 8,700 examples in nine languages including four African languages (Hausa, Igbo, Swahili, and Yorùbá) and a zero-shot test language (Russian). We additionally release screenshots of the original figures for future research on multilingual multi-modal approaches. Through an in-depth human evaluation, we show that TaTa is challenging for current models and that less than half the outputs from an mT5-XXL-based model are understandable and attributable to the source data. We further demonstrate that existing metrics perform poorly for TaTa and introduce learned metrics that achieve a high correlation with human judgments. We release all data and annotations at https://github.com/google-research/url-nlp.