Boming Yang

Ph.D Candidate Student at UTokyo, TMI, Matsuo Lab

Education

  • Ph.D in Technology Management for Innovation, The University of Tokyo (expected)
  • M.S in Information Science & Technology, The University of Tokyo
  • B.S in Computer Secience & Technology, The University of Science and Technology Beijing

Pulications

Rui Yang, Boming Yang, Sixun Ouyang, Tianwei She, Aosong Feng, Yuang Jiang, Freddy Lecue, Jinghui Lu, Irene Li: Leveraging Large Language Models for Concept Graph Recovery and Question Answering in NLP Education (arXiv)

Dairui Liu, Boming Yang, Honghui Du, Derek Greene, Aonghus Lawlor, Ruihai Dong, Irene Li: RecPrompt: A Prompt Tuning Framework for News Recommendation Using Large Language Models (arXiv)

Boming Yang, Dairui Liu, Toyotaro Suzumura, Ruihai Dong and Irene Li: Going Beyond Local: Global Graph‐Enhanced Personalized News Recommendations.(RecSys 2023, Best Student Paper)

Zihui Li and Boming Yang: NNKGC: Improving Knowledge Graph Completion with Node Neighborhoods (DL4KG workshop, ISWC 2023)

Reasearch experience

  • Nikkei Collaboration: Learning User Representation for Downstream Tasks (Oct. 2022 - Now)
    • The University of Tokyo, Tokyo
    • Utilized contrastive learning and data augmentation techniques to learn user interest representation from historical reading sequences.
    • Applied NLP methods like pre‐trained language model and graph methods to extract news content information for more accurate modeling of news with attention mechanisms.
    • Supervisor: Prof. Toyotaro Suzumura
  • R-CCS: Forecasting the future imports and exports of Japan (Apr. 2023 - Now)
    • RIKEN Center for Computational Science, Tokyo
    • Conducted thorough research to identify potential data sources and performed comprehensive data preprocessing to extract valuable datasets.
    • Employed advanced deep time-series techniques and traditional methods for accurate forecasting, while also exploring the implementation of Ensemble Learning to capture the combined insights from multiple models.
    • Supervisor: Prof. Maddegedara Lalith
  • GRAPE: Parallelizing sequential graph computations (Mar. 2017 - Mar. 2018):
    • Research Assistant, Beihang University, Beijing
    • Restructured the classical algorithm SSSP,PageRank,VF2,etc. on GRAPE system,as well as passed th correctness verification of big datatest in Milky Way‐2 supercomputer.
    • Made significant contributions to the GRAPE team’s achievement of the ACMSIGMOD’17 Best Paper Award and VLDB’17 Best Demo Award.
    • Supervisor: Prof. Wenfei Fan

Professional experience

  • SmartNews Co.m Ltd., Japan (Nov. 2023 - Now)
    • Research Intern
    • Implemented and optimized keyphrase extraction algorithms to enhance the accuracy of news categorization and improve user content discovery experience.
  • BestSkip Co., Ltd., Japan (Apr. 2021 - Mar. 2022)
    • Algorithm Engineer
    • Developed and optimized a real‐time location algorithm using both Java and C,resulting in improved calculation speed and increased signal processing capabilities.
    • Invented auxiliary methods to overcome indoor positioning height instability and used filters to dynamically control the smoothness of the trajectory.
  • Fma (Bejing) Technology Co., Ltd., China (Mar. 2018 - Aug. 2019):
    • Software Engineering Intern
    • Cooperated with Sina to analyze social network graphs,and improve the algorithm to achieve it in a lighter environment and a shorter time.
    • Restructured the subgraph isomorphism algorithm VF2 and TurboISO into parallel and distributed versions on the graph computation engine PandaGraph.

Honors & Awards

  • 2023 Recsys 2023, Best Student Paper.
  • 2019 Outstanding Graduate honor (5%), Univericity‐level
  • 2018 International Second Prize, ASC 18 Student Supercomputer Challenge
  • 2017 Group Second Prize, China Collegiate Computing Contest ‐ Group Programming Ladder Tournament
  • 2017 National First Prize scholarship (5%) , Univericity‐level