We introduce UPAIR, a novel human-annotated benchmark for evaluating multimodal image retrieval in the context of utility patents. UPAIR addresses current limitations by providing: Three-dimensiona
The dataset used in the paper is a text and image dataset, with 5477 instances, 3201 attributes, and 248 categories. The dataset is used for clustering and evaluation of the proposed method.
The Patents dataset includes information about Patents citing Springer Nature publications. See also: https://scigraph.springernature.com/explorer/datasets/data_at_a_glance/ NOTE: this datas
