• Publications
  • Influence
Deep Graph Contrastive Representation Learning
This paper proposes a novel framework for unsupervised graph representation learning by leveraging a contrastive objective at the node level, and generates two graph views by corruption and learns node representations by maximizing the agreement of node representations in these two views.
Graph Contrastive Learning with Adaptive Augmentation
This paper proposes a novel graph contrastive representation learning method with adaptive augmentation that incorporates various priors for topological and semantic aspects of the graph that consistently outperforms existing state-of-the-art baselines and even surpasses some supervised counterparts.
On the Recurrent Ulva prolifera Blooms in the Yellow Sea and East China Sea
[1] A massive bloom of the green macroalgae Ulva prolifera (previously known as Enteromorpha prolifera) occurred in June 2008 in the Yellow Sea (YS), resulting in perhaps the largest “green tide”
TAGNN: Target Attentive Graph Neural Networks for Session-based Recommendation
In TAGNN, target-aware attention adaptively activates different user interests with respect to varied target items and the learned interest representation vector varies with different target items, greatly improving the expressiveness of the model.
Both Boceprevir and GC376 efficaciously inhibit SARS-CoV-2 by targeting its main protease
  • L. Fu, F. Ye, +13 authors G. Gao
  • Biology, Medicine
    Nature communications
  • 4 September 2020
It is shown that a clinically approved anti-HCV drug and a pre-clinical inhibitor against feline infectious peritonitis (corona) virus (FIPV), GC376, both efficaciously inhibit SARS-CoV-2 in Vero cells by targeting Mpro.
Creep behavior of a zirconium diboride–silicon carbide composite
Abstract Flexural creep studies of ZrB 2 –20 vol% SiC ultra-high temperature ceramic were conducted over the range of 1400–1820 °C in an argon shielded testing apparatus. A two decade increase in
Algorithms for channel assignment in mobile wireless networks using temporal coloring
This work analyzes two simple algorithms called SNAP and SMASH that take diametrically opposite positions on colors vs re-assignments, and provides theoretical results on the ranges of α in which one outperforms the other, both for arbitrary and random time-varying graphs.
Restriction landmark genome scanning for aberrant methylation in primary refractory and relapsed acute myeloid leukemia; involvement of the WIT-1 gene
The results indicate that RLGS-M can be used to find novel epigenetic alterations in human cancer that are undetectable by standard methods and underline the potential importance of WIT-1 methylation in chemoresistant AML.
Activation of Ca2+-sensing receptor as a protective pathway to reduce Cadmium-induced cytotoxicity in renal proximal tubular cells
The identification of the activation of CaSR-mediated protective pathway in renal cells sheds light on a possible cellular protective mechanism against Cd-induced kidney injury.
Comparative proteomic analysis of rat hepatic stellate cell activation: A comprehensive view and suppressed immune response
Bioinformatics analyses revealed that the 319 up‐regulated proteins represented multiple cellular functions closely associated with HSC activation, such as extracellular matrix synthesis and proliferation, and the top biological function, top network, and top canonical pathway of down‐ regulated proteins were all involved in immune responses.