Thrombosis and inflammation are intricately linked in several major clinical disorders, including disseminated intravascular coagulation and acute ischemic events. The damage-associated molecular pattern molecule high-mobility group box 1 (HMGB1) is upregulated by activated platelets in multiple inflammatory diseases; however, the contribution of platelet-derived HMGB1 in thrombosis remains unexplored. Here, we generated transgenic mice with platelet-specific ablation of HMGB1 and determined that platelet-derived HMGB1 is a critical mediator of thrombosis. Mice lacking HMGB1 in platelets exhibited increased bleeding times as well as reduced thrombus formation, platelet aggregation, inflammation, and organ damage during experimental trauma/hemorrhagic shock. Platelets were the major source of HMGB1 within thrombi. In trauma patients, HMGB1 expression on the surface of circulating platelets was markedly upregulated. Moreover, evaluation of isolated platelets revealed that HMGB1 is critical for regulating platelet activation, granule secretion, adhesion, and spreading. These effects were mediated via TLR4- and MyD88-dependent recruitment of platelet guanylyl cyclase (GC) toward the plasma membrane, followed by MyD88/GC complex formation and activation of the cGMP-dependent protein kinase I (cGKI). Thus, we establish platelet-derived HMGB1 as an important mediator of thrombosis and identify a HMGB1-driven link between MyD88 and GC/cGKI in platelets. Additionally, these findings suggest a potential therapeutic target for patients sustaining trauma and other inflammatory disorders associated with abnormal coagulation.
Industrial recommender systems usually consist of the matching stage and the ranking stage, in order to handle the billion-scale of users and items. The matching stage retrieves candidate items relevant to user interests, while the ranking stage sorts candidate items by user interests. Thus, the most critical ability is to model and represent user interests for either stage. Most of the existing deep learning-based models represent one user as a single vector which is insufficient to capture the varying nature of user's interests. In this paper, we approach this problem from a different view, to represent one user with multiple vectors encoding the different aspects of the user's interests. We propose the Multi-Interest Network with Dynamic routing (MIND) for dealing with user's diverse interests in the matching stage. Specifically, we design a multi-interest extractor layer based on capsule routing mechanism, which is applicable for clustering historical behaviors and extracting diverse interests. Furthermore, we develop a technique named label-aware attention to help learn a user representation with multiple vectors. Through extensive experiments on several public benchmarks and one largescale industrial dataset from Tmall, we demonstrate that MIND can achieve superior performance than state-of-the-art methods for recommendation. Currently, MIND has been deployed for handling major online traffic at the homepage on Mobile Tmall App.
Thermally stable Au single-atoms supported by monolayered CuO grown at Cu(110) have been successfully prepared. The charge transfer from the CuO support to single Au atoms is confirmed to play a key role in tuning the activity for CO oxidation. Initially, the negatively charged Au single-atom is active for CO oxidation with its adjacent lattice O atom depleted to generate an O vacancy in the CuO monolayer. Afterward, the Au single-atom is neutralized, preventing further CO reaction. The produced O vacancy can be healed by exposure to O at 400 K and accordingly the reaction activity is restored.
Deep learning based methods have been widely used in industrial recommendation systems (RSs). Previous works adopt an Em-bedding&MLP paradigm: raw features are embedded into lowdimensional vectors, which are then fed on to MLP for final recommendations. However, most of these works just concatenate different features, ignoring the sequential nature of users' behaviors. In this paper, we propose to use the powerful Transformer model to capture the sequential signals underlying users' behavior sequences for recommendation in Alibaba. Experimental results demonstrate the superiority of the proposed model, which is then deployed online at Taobao and obtain significant improvements in online Click-Through-Rate (CTR) comparing to two baselines.
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.