Automated visual understanding of our diverse and open world demands computer vision models to generalize well with minimal customization for specific tasks, similar to human vision. Computer vision foundation models, which are trained on diverse, large-scale dataset and can be adapted to a wide range of downstream tasks, are critical for this mission to solve real-world computer vision applications. While existing vision foundation models such as CLIP (Radford et al., 2021), ALIGN (Jia et al., 2021), and Wu Dao 2.0 (Wud) focus mainly on mapping images and textual representations to a cross-modal shared representation, we introduce a new computer vision foundation model, Florence, to expand the representations from coarse (scene) to fine (object), from static (images) to dynamic (videos), and from RGB to multiple modalities (caption, depth). By incorporating universal visual-language representations from Web-scale image-text data, our Florence model can be easily adapted for various computer vision tasks, such as classification, retrieval, object detection, VQA, image caption, video retrieval and action recognition. Moreover,

Developing high‐efficiency electrocatalysts for the hydrogen evolution and oxidation reactions (HER/HOR) in alkaline electrolytes is of critical importance for realizing renewable hydrogen technologies. Ruthenium phosphides (RuPx) are promising candidates to substitute Pt‐based electrodes; however, great challenges still remain in their electronic structure regulation for optimizing intermediate adsorption. Herein, it is reported that a homologous RuP@RuP2 core–shell architecture constructed by a phosphatization‐controlled phase‐transformation strategy enables strong electron coupling for optimal intermediate adsorption by virtue of the emergent interfacial functionality. Density functional theory calculations show that the RuP core and RuP2 shell present efficient electron transfer, leading to a close to thermoneutral hydrogen adsorption Gibbs free energy of 0.04 eV. Impressively, the resulting material exhibits superior HER/HOR activities in alkaline media, which outperform the benchmark Pt/C and are among the best reported bifunctional hydrogen electrocatalysts. The present work not only provides an efficient and cost‐effective bifunctional hydrogen electrocatalyst, but also offers a new synthetic protocol to rationally synthesize homologous core–shell nanostructures for widespread applications.

No abstract

BackgroundThe coiled-coil domain containing (CCDC) family proteins have important biological functions in various diseases. However, the coiled-coil domain containing 137 (CCDC137) was rarely studied. We aim to investigate the role of CCDC137 in pan-cancer.MethodsCCDC137 expression was evaluated in RNA sequence expression profilers of pan-cancer and normal tissues from The Cancer Genome Atlas (TCGA) and Genotype-Tissue Expression (GTEx) database. The influence of CCDC137 on the prognosis of tumor patients was analyzed using clinical survival data from TCGA. Function and pathway enrichment analysis was performed to explore the role of CCDC137 using the R package “clusterProfiler.” We further analyzed the correlation of immune cell infiltration score of TCGA samples and CCDC137 expression using TIMER2 online database.ResultsCCDC137 was over-expressed and associated with worse survival status in various tumor types. CCDC137 expression was positively correlated with tumor associated macrophages (TAMs) and cancer associated fibroblasts (CAFs) in Lower Grade Glioma (LGG) and Uveal Melanoma (UVM). In addition, high CCDC137 expression was positively correlated with most immunosuppressive genes, including TGFB1, PD-L1, and IL10RB in LGG and UVM.ConclusionsOur study identified CCDC137 as an oncogene and predictor of worse survival in most tumor types. High CCDC137 may contribute to elevated infiltration of TAMs and CAFs and be associated with tumor immunosuppressive status.

No abstract