Crossmodal-3600: A Massively Multilingual Multimodal Evaluation Dataset
This addresses the problem of evaluating multilingual image captioning models for researchers, though it is incremental as it focuses on dataset creation rather than a new method.
The paper tackles the lack of high-quality evaluation datasets for massively multilingual image captioning by introducing Crossmodal-3600, a dataset of 3600 images with human-generated captions in 36 languages, and shows it provides superior correlation with human evaluations for model selection.
Research in massively multilingual image captioning has been severely hampered by a lack of high-quality evaluation datasets. In this paper we present the Crossmodal-3600 dataset (XM3600 in short), a geographically diverse set of 3600 images annotated with human-generated reference captions in 36 languages. The images were selected from across the world, covering regions where the 36 languages are spoken, and annotated with captions that achieve consistency in terms of style across all languages, while avoiding annotation artifacts due to direct translation. We apply this benchmark to model selection for massively multilingual image captioning models, and show superior correlation results with human evaluations when using XM3600 as golden references for automatic metrics.