XLXubin Li

Xubin Li

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Xubin Li has a total of 60 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Accelerating Image Generation with Sub-path Linear Approximation Model

    by Chen Xu, Tianhui Song, Weixin Feng, Xubin Li, Tiezheng Ge, Bo Zheng, Limin Wang


  • Open Access

    BOMGraph: Boosting Multi-scenario E-commerce Search with a Unified Graph Neural Network

    by Shuai Fan, Jinping Gou, Yang Li, Jiaxing Bai, Chen Lin, Wanxian Guan, Xubin Li, Hongbo Deng, Jian Xu, Bo Zheng

    Mobile Taobao Application delivers search services on multiple scenarios that take textual, visual, or product queries. This paper aims to propose a unified graph neural network for these search scenarios to leverage data from multiple scenarios and jointly optimize search performances with less training and maintenance costs. Towards this end, this paper proposes BOMGraph, BOosting Multi-scenario E-commerce Search with a unified Graph neural network. BOMGraph is embodied with several components to address challenges in multi-scenario search. It captures heterogeneous information flow across scenarios by inter-scenario and intra-scenario metapaths. It learns robust item representations by disentangling specific characteristics for different scenarios and encoding common knowledge across scenarios. It alleviates label scarcity and long-tail problems in scenarios with low traffic by contrastive learning with cross-scenario augmentation. BOMGraph has been deployed in production by Alibaba's E-commerce search advertising platform. Both offline evaluations and online A/B tests demonstrate the effectiveness of BOMGraph.


  • Open Access

    E-commerce Search via Content Collaborative Graph Neural Network

    by Guipeng Xv, Chen Lin, Wanxian Guan, Jinping Gou, Xubin Li, Hongbo Deng, Jian Xu, Bo Zheng

    Recently, many E-commerce search models are based on Graph Neural Networks (GNNs). Despite their promising performances, they are (1) lacking proper semantic representation of product contents; (2) less efficient for industry-scale graphs; and (3) less accurate on long-tail queries and cold-start products. To address these problems simultaneously, this paper proposes CC-GNN, a novel Content Collaborative Graph Neural Network. Firstly, CC-GNN enables content phrases to participate explicitly in graph propagation to capture the proper meaning of phrases and semantic drifts. Secondly, CC-GNN presents several efforts towards a more scalable graph learning framework, including efficient graph construction, MetaPath-guided Message Passing, and Difficulty-aware Representation Perturbation for graph contrastive learning. Furthermore, CC-GNN adopts Counterfactual Data Supplement at both supervised and contrastive learning to resolve the long-tail/cold-start problems. Extensive experiments on a real E-commerce dataset of 100-million-scale nodes show that CC-GNN produces significant improvements over existing methods (i.e., more than 10% improvements in terms of several key evaluation metrics for overall, long-tail queries and cold-start products) while reducing computational complexity. The proposed components of CC-GNN can be applied to other models for search and recommendation tasks. Experiments on a public dataset show that applying the proposed components can improve the performance of different recommendation models.


  • Open Access

    Multi-Scenario Ranking with Adaptive Feature Learning

    by Yu Tian, Bofang Li, Si Chen, Xubin Li, Hongbo Deng, Jian Xu, Bo Zheng, Qian Wang, Chenliang Li


  • Computed tomography–based radiomics machine learning classifiers to differentiate type I and type II epithelial ovarian cancers

    by Jiaojiao Li, Xubin Li, Juanwei Ma, Fang Wang, Shujun Cui, Zhaoxiang Ye

    Objectives To compare computed tomography (CT)–based radiomics for preoperatively differentiating type I and II epithelial ovarian cancers (EOCs) using different machine learning classifiers and to construct and validate the best diagnostic model. Methods A total of 470 patients with EOCs were included retrospectively. Patients were divided into a training dataset (N = 329) and a test dataset (N = 141). A total of 1316 radiomics features were extracted from the portal venous phase of contrast-enhanced CT images for each patient, followed by dimension reduction of the features. The support vector machine (SVM), k-nearest neighbor (KNN), random forest (RF), naïve Bayes (NB), logistic regression (LR), and eXtreme Gradient Boosting (XGBoost) classifiers were trained to obtain the radiomics signatures. The performance of each radiomics signature was evaluated and compared by the area under the receiver operating characteristic curve (AUC) and relative standard deviation (RSD). The best radiomics signature was selected and combined with clinical and radiological features to establish a combined model. The diagnostic value of the combined model was assessed. Results The LR-based radiomics signature performed well in the test dataset, with an AUC of 0.879 and an accuracy of 0.773. The combined model performed best in both the training and test datasets, with AUCs of 0.900 and 0.934 and accuracies of 0.848 and 0.823, respectively. Conclusion The combined model showed the best diagnostic performance for distinguishing between type I and II EOCs preoperatively. Therefore, it can be a useful tool for clinical individualized EOC classification. Key Points • Radiomics features extracted from computed tomography (CT) could be used to differentiate type I and II epithelial ovarian cancers (EOCs). • Machine learning can improve the performance of differentiating type I and II EOCs. • The combined model exhibited the best diagnostic capability over the other models in both the training and test datasets.


  • Study on Cavity Evolution of AZ80 Magnesium Alloy During Rotary Extrusion

    by Xubin Li, Qiang Wang, Qingling Fang, Zhimin Zhang

  • Open Access

    Abstract 1002: Targeting adaptation to cancer treatment by drug combinations

    by Heping Wang, Augustin Luna, Gonghong Yan, Xubin Li, Ozgun Babur, Chris Sander, Gordon Mills

    Adaptation of tumors to therapeutic interventions contributes to dismal long-term patient outcomes. Adaptation to therapy involves the co-action of functionally related proteins that together activate cell survival programs and compensate for the therapeutic impact. Oncogenic dependencies to such adaptive events, however, can generate new therapeutic vulnerabilities that can be targeted with drug combinations. The precision medicine approaches in which targeted drugs are matched to pre-existing genomic aberrations fail to address the adaptive responses and resulting vulnerabilities. Here, we provide the mathematical formulation, implementation, and validation of the TargetScore method. The TargetScore identifies collective adaptive responses to targeted interventions as concurrent changes of phospho-proteins that are connected within a signaling network. Based on the adaptive responses, the method predicts drug-induced vulnerabilities. Using TargetScore, we inferred the adaptive responses with short-term (i.e., days) stress and long-term (i.e., months) acquired resistance to inhibitors of antiapoptotic mediators, MCL1 and BCL2. With experiments guided by the predictions, we identified synergistic interactions between inhibitors of PARP, SHP2, and MCL1 in breast cancer cells. TargetScore is readily applicable to existing precision oncology efforts by matching targeted drug combinations to emerging molecular signatures under therapeutic stress. Citation Format: Heping Wang, Augustin Luna, Gonghong Yan, Xubin Li, Ozgun Babur, Chris Sander, Gordon Mills. Targeting adaptation to cancer treatment by drug combinations [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1002.

  • Open Access

    The Microstructure Evolution of Mg-RE Alloy Produced by Reciprocating Upsetting Extrusion during Hot Compression

    by Ziwei Zhang, Jianmin Yu, Zeru Wu, Hongbing Hu, Zhimin Zhang, Mo Meng, Yong Xue, Xubin Li

    Mg-13Gd-4Y-2Zn-0.4Zr (wt. %) alloy bar produced by three passes reciprocating upsetting extrusion (named as RUE-ed bar) exhibited fine grain with the average grain size of 3.02 μm. Hot compression tests of the RUE-ed bar were carried out on Gleeble-3800 compression unit at different deformation temperatures (653, 683, 713, and 743 K) and strain rates (0.001–1 s, 0.01–1 s, 0.1–1 s, and 0.5–1 s). This alloy showed work hardening and softening stages in hot compression, the thermal activation energy of the RUE-ed bar was 150 ± 1 kJ/mol and the constitutive equation was: ε˙=1.80×109[sinh(0.0174σ)]2.47exp[−150×1038.314×T]. Numerous Mg5 (Gd, Y, Zn) phase re-dissolved in α-Mg matrix appeared in the RUE-ed samples during hot compression deformation. The movement of the dislocation stimulated the re-dissolution of the Mg5 (Gd, Y, Zn) phase. The re-dissolution of Mg5 (Gd, Y, Zn) phase promoted texture strengthening and DRX grains growth in this experiment. In addition, the transformation and kinking of LPSO phase played an important coordinating role in the process of hot compression; 18R-LPSO was changed to 14H-LPSO phase at low strain rate while the LPSO phase kinked dominant to coordinated deformation at high strain rate.


  • Open Access

    Simulation and Experimental Study on Roll-Forming Limit of Cup

    by Baohong Zhang, Shengpeng Ning, Zeng Wei, Yali Duan, Xubin Li, Yongbiao Yang

    Roll forming can improve the material utilization rate and production efficiency of cups with a curved rotary profile, but there is no basis for the determination of forming limit. The DEFORM-3D software was used to simulate the roll forming of cups. The influence of the billet wall thickness and bottom thickness, coefficient of friction, radius of roller, and the fillet radius of the punch on the forming limit was studied, and the damage value and velocity vector were analyzed. The results showed that the forming limit of the billet’s wall thickness in roll forming for a cup is about 62%. With the increase of the ratio of the formed cup’s wall thickness to the billet’s bottom thickness, the forming limit of wall thickness will be slightly reduced. A larger radius of roller, fillet radius of punch, and friction coefficient between punch and billet and a smaller friction coefficient between roller and billet are good for decreasing the damage value and improving the roll-forming limit. According to the numerical simulation results, the roll-forming limit diagram of cups is established, and the accuracy of the forming limit diagram is verified by experiments.