The goal of WMT 2018 Shared Task on Translation Quality Estimation is to investigate automatic methods for estimating the quality of machine translation results without reference translations. This paper presents the QE Brain system, which proposes the neural Bilingual Expert model as a feature extractor based on conditional target language model with a bidirectional transformer and then processes the semantic representations of source and the translation output with a Bi-LSTM predictive model for automatic quality estimation. The system has been applied to the sentence-level scoring and ranking tasks as well as the wordlevel tasks for finding errors for each word in translations. An extensive set of experimental results have shown that our system outperformed the best results in WMT 2017 Quality Estimation tasks and obtained top results in WMT 2018. * * indicates equal contribution.
The performances of machine translation (MT) systems are usually evaluated by the metric BLEU when the golden references are provided. However, in the case of model inference or production deployment, golden references are usually expensively available, such as human annotation with bilingual expertise. In order to address the issue of translation quality estimation (QE) without reference, we propose a general framework for automatic evaluation of the translation output for the QE task in the Conference on Statistical Machine Translation (WMT). We first build a conditional target language model with a novel bidirectional transformer, named neural bilingual expert model, which is pre-trained on large parallel corpora for feature extraction. For QE inference, the bilingual expert model can simultaneously produce the joint latent representation between the source and the translation, and real-valued measurements of possible erroneous tokens based on the prior knowledge learned from parallel data. Subsequently, the features will further be fed into a simple Bi-LSTM predictive model for quality estimation.
BackgroundThe exact biological role of PCOLCE was not yet clear and there were few reports study the correlation of PCOLCE gene expression level with the occurrence and development of gastric cancer.MethodsThe expression of PCOLCE was analyzed by performing the Oncomine and Ualcan database. We evaluated the function of PCOLCE on clinical prognosis with the use of Kaplan–Meier plotter database. The relationship between PCOLCE and cancer immune in filtrates was researched by Tumor Immune Estimation Resource (TIMER) site database.ResultsPCOLCE significantly upregulated in gastric cancer patients compared to normal gastric samples. And the increased expression of PCOLCE mRNA was closely linked to shorter overall survival (OS), progress-free survival (PFS) in all gastric cancers. Besides, PCOLCE expression displayed a tight correlation with infiltrating levels of macrophages and dendritic cells (DCs) in gastric cancer. Moreover, PCOLCE expression was positively correlated with diverse immune marker sets in gastric cancer.ConclusionAll the results above suggested that overexpression of PCOLCE indicated unfavorable prognosis in patients with gastric cancer. PCOLCE was correlated with immune infiltrating levels including those of B cells, CD8 + T cells, CD4 + T cells, macrophages, neutrophils, and DCs in gastric cancer patients. All the findings suggested that PCOLCE could be used as a prognostic biomarker for determining prognosis and immune infiltration in gastric cancer. Additionally, PCOLCE expression potentially contributed to the regulation of monocyte, M2 macrophage, Tfh, CD8 + T cell, TAM, Th1 cell Thus PCOLCE is a potential target for gastric cancer therapy and these preliminary findings require further study to determine whether PCOLCE-targeting reagents might be developed for clinical application in gastric cancer.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.