JZJunsheng Zhang

Junsheng Zhang

*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


Junsheng Zhang has a total of 88 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Identifying multidisciplinary problems from scientific publications based on a text generation method

    by Ziyan Xu, Hongqi Han, Linna Li, Junsheng Zhang, Zexu Zhou

    Purpose: A text generation based multidisciplinary problem identification method is proposed, which does not rely on a large amount of data annotation. Design/methodology/approach: The proposed method first identifies the research objective types and disciplinary labels of papers using a text classification technique; second, it generates abstractive titles for each paper based on abstract and research objective types using a generative pre-trained language model; third, it extracts problem phrases from generated titles according to regular expression rules; fourth, it creates problem relation networks and identifies the same problems by exploiting a weighted community detection algorithm; finally, it identifies multidisciplinary problems based on the disciplinary labels of papers. Findings: Experiments in the “Carbon Peaking and Carbon Neutrality” field show that the proposed method can effectively identify multidisciplinary research problems. The disciplinary distribution of the identified problems is consistent with our understanding of multidisciplinary collaboration in the field. Research limitations: It is necessary to use the proposed method in other multidisciplinary fields to validate its effectiveness. Practical implications: Multidisciplinary problem identification helps to gather multidisciplinary forces to solve complex real-world problems for the governments, fund valuable multidisciplinary problems for research management authorities, and borrow ideas from other disciplines for researchers. Originality/value: This approach proposes a novel multidisciplinary problem identification method based on text generation, which identifies multidisciplinary problems based on generative abstractive titles of papers without data annotation required by standard sequence labeling techniques.


  • Open Access

    No semistability at infinity for Calabi-Yau metrics asymptotic to cones

    by Song Sun, Junsheng Zhang

    We discover a "no semistability at infinity" phenomenon for complete Calabi-Yau metrics asymptotic to cones, by eliminating the possible appearance of an intermediate K-semistable cone in the 2-step degeneration theory developed by Donaldson and the first author. It is in sharp contrast to the setting of local singularities of K\"ahler-Einstein metrics. A byproduct of the proof is a polynomial convergence rate to the asymptotic cone for such manifolds, which bridges the gap between the general theory of Colding-Minicozzi and the classification results of Conlon-Hein.

  • Open Access

    Low-dose apatinib combined with neoadjuvant chemotherapy in the treatment of early-stage triple-negative breast cancer (LANCET): a single-center, single-arm, phase II trial

    by Ciqiu Yang, Junsheng Zhang, Yi Zhang, Fei Ji, Yitian Chen, Teng Zhu, Liulu Zhang, Hongfei Gao, Mei Yang, Jieqing Li, et al.

    Background: Antiangiogenic therapy combined with chemotherapy could improve pathological complete response (pCR) for breast cancer. Apatinib is an oral tyrosine kinase inhibitor that selectively inhibits vascular endothelial growth factor receptor 2. We assessed the efficacy and safety of apatinib combined with standard neoadjuvant chemotherapy in patients with triple-negative breast cancer (TNBC). Materials and methods: This single-arm, phase II study enrolled patients aged 18–70 years with previously untreated stage IIA-IIIB TNBC. Patients received oral apatinib at a dose of 250 mg once daily and intravenously docetaxel every 3 weeks for four cycles, followed by epirubicin plus cyclophosphamide every 3 weeks for four cycles. The primary endpoint was the pCR rate in the breast and lymph nodes. Secondary endpoints included objective response rate, event-free survival (EFS), overall survival (OS), and safety. Results: In all, 31 patients were enrolled, and the median follow-up time was 22.9 months (range: 10.1–41.6 months). The pCRs in both breast and lymph nodes were achieved in 17 [54.8%; 95% confidence interval (CI): 36.0–72.7] of 31 patients. Objective responses were achieved in 29 patients (93.5%; 95% CI: 78.6–99.2), and disease control was achieved in 31 patients (100%; 95% CI: 88.8–100.0). The 2-year EFS and 2-year OS were 90.9% and 94.4%, respectively. The five most common treatment-related adverse events were fatigue (51%), hypertension (41%), anorexia (39%), hand–foot syndrome (35%), and diarrhea (32%). Few grade 3 or more adverse events were observed. Conclusion: The combination of apatinib with docetaxel followed by epirubicin plus cyclophosphamide showed excellent efficacy and manageable toxicities; and further randomized controlled phase III trials are warranted. Trial registration: This trial was registered with ClinicalTrials.gov (NCT03243838) on 5 August 2017.


  • Open Access

    Sequencing and analysis of the complete mitochondrial genome of the lesser bandicoot rat (Bandicota bengalensis) from China and its phylogenetic analysis

    by Liu Zhu, Wang Qing-Qing, Jiang Xue-Ting, Jiang Wen-Jing, Zhao Xin-Xu, Yao Qian-Qian, Zhang Jun-Sheng, Piao Zhong-Wan

    The complete mitogenome sequence of the lesser bandicoot rat (Bandicota bengalensis Gray and Hardwicke, 1833) was determined using long PCR. The genome was 16,327 bp in length and contained 13 protein-coding genes, 2 ribosomal RNA genes, 22 transfer RNA genes, 1 origin of L strand replication and 1 control region. The overall base composition of the heavy strand is A (34.2%), C (24.9%), T (28.5%) and G (12.4%). The base compositions present clearly the A–T skew, which is most obviously in the control region and protein-coding genes. Mitochondrial genome analyses based on MP, ML, NJ and Bayesian analyses yielded identical phylogenetic trees. This study verifies the evolutionary status of Bandicota bengalensis in Muridae at the molecular level. The mitochondrial genome would be a significant supplement for the Bandicota bengalensis genetic background. The two Bandicota species formed a monophyletic group with the high bootstrap value (100%) in all examinations.


  • Analyzing online impression management ability of Chinese teenagers

    by Zengquan Fang, Xiaoxu Ji, Xuejing Qi, Junsheng Zhang


  • Open Access

    GSA-Net: gated scaled dot-product attention based neural network for reading comprehension

    by Xiang Ma, Junsheng Zhang

    Reading Comprehension (RC) is concerned with building systems that automatically answer questions about a given context passage. The interactions between the context and question are very important to locate the correct answer. In this paper, we propose a Gated Scaled Dot-Product Attention based model for RC task. The character-level embedding is incorporated into the word embedding which is helpful to deal with Out-of-Vocabulary (OOV) tokens. The attention distribution is obtained by scaled dot product which captures the interaction between question and passage effectively. Further, self-matching attention mechanism is adopted to resolve the problem of long-distance dependency. These components provides more information for the prediction of the starting and ending position of the answer. We evaluate our method on Stanford Question Answering Dataset (SQuAD) and the results show that different components in the model boost the performance.


  • Open Access

    A Heterogeneous Battlefield Situation Information Sharing Method Based on Content

    by Cuntao Liu, Wendong Zhao, Aijing Li, Junsheng Zhang

    Various information systems adopt different information category standards and description methods even for the same battlefield space. This leads to the heterogeneity of information distributed in different systems, which hinders the information sharing among different systems. In this paper, we adopt the idea of schema mapping and design a framework of realizing heterogeneous information sharing based on content. We design a concept-logic tree model to organize the battlefield situation information and realize the mapping from local concept models in different systems to a global unified concept model. By constructing a unified information space, we realize the centralized organization, storage, and management of entity description information. Then, the information broadcasting mechanism and content based information query mechanism are designed to realize information sharing. Theoretical analysis and experiment results verify the effectiveness of the proposed framework.


  • Open Access

    Event-centric Machine Strategic Reading Approach for Scientific Literature

    by Junsheng Zhang, Yunchuan Sun, Yuan Gao, Xinyue Fang