Evaluating image captions is very challenging partially due to the fact that there are multiple correct captions for every single image. Most of the existing one-to-one metrics operate by penalizing mismatches between reference and generative caption without considering the intrinsic variance between ground truth captions. It usually leads to over-penalization and thus a bad correlation to human judgment. Recently, the latest one-to-one metric BERTScore can achieve high human correlation in system-level tasks while some issues can be fixed for better performance. In this paper, we propose a novel metric based on BERTScore that could handle such a challenge and extend BERTScore with a few new features appropriately for image captioning evaluation. The experimental results show that our metric achieves state-of-the-art human judgment correlation.

No abstract

Accurately‐labeled data is crucial for the training of machine learning models. For singing‐related tasks in the music information retrieval field, accurately‐labeled data is limited because annotating singing is time‐consuming. Several studies create vocal datasets using a two‐step annotation method which creates coarse labels first and then executes a manual calibration procedure. However, manually calibrating coarsely‐labeled singing data is expensive and time‐consuming. To address this problem, in this study we propose a singing‐label calibration framework, which aims to automatically calibrate the coarsely‐labeled singing data with higher accuracy. This framework contains a data augmentation method to generate training and testing data, a reasonable data preprocessing method to handle music audio and symbolic labels, a fully‐convolutional neural network to estimate the difference between coarse labels and accurate labels, and a novel calibration function to correct the coarse labels. Various experiments are conducted to examine the effect of our research. The results show that our model can highly reduce the cost time and slightly increase the labeling accuracy of the manual calibration process. © 2023 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC.

No abstract

In the detection of protein complexes, the completeness of a protein–protein interaction (PPI) network is crucial. Complete PPI networks, however, are not available for most species because experimentally identified PPIs are usually very limited. This paper proposes a deep learning based PPI predictor to construct a complete PPI network, from which protein complexes are detected using a spectral clustering method. For this purpose, the unknown PPIs are estimated by using a deep PPI predictor consisting of a semi‐supervised SVM classifier and a deep feature extractor of the convolutional neural network (CNN). Meanwhile, the similarities of gene ontology (GO) annotations contribute to protein interactions, and the differences of subcellular localizations contribute to negative interactions. Considering that, we pretrain the deep CNN feature extractor in a class of deep GO annotation and subcellular localization predictors using datasets from the type species, then transfer it to the PPI prediction model for fine‐tuning. In this way, we have a deep PPI detector enhanced with transfer learning of GO annotation and subcellular localization prediction. Experimental results on benchmark datasets show that the proposed method outperforms the state‐of‐the‐art methods. © 2021 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC.