PatFig: Generating Short and Long Captions for Patent Figures
This addresses the need for automated captioning in patent analysis, but it is incremental as it builds on existing LVLM methods with a new dataset.
The paper tackled the problem of generating captions for patent figures by introducing a large-scale dataset with 30,000+ figures and multiple annotations, and they finetuned an LVLM model to produce short and long descriptions, achieving results that demonstrate the dataset's usability.
This paper introduces Qatent PatFig, a novel large-scale patent figure dataset comprising 30,000+ patent figures from over 11,000 European patent applications. For each figure, this dataset provides short and long captions, reference numerals, their corresponding terms, and the minimal claim set that describes the interactions between the components of the image. To assess the usability of the dataset, we finetune an LVLM model on Qatent PatFig to generate short and long descriptions, and we investigate the effects of incorporating various text-based cues at the prediction stage of the patent figure captioning process.