avatar

Alex (Yuanming) Tao

PhD student in applied maths
The University of Sydney
ytao5219 (at) gmail (dot) com

[Google Scholar] [Github] [LinkedIn] [Twitter]

Short Bio

I am a PhD student in Applied Mathematics at The University of Sydney, working under the supervision of Prof. Eduardo G. Altman and Prof. Tristram Alexander. My research focuses on developing methodological approaches to uncover meso-scale structures in complex networks.

I graduated with first-class honours in Applied Mathematics from The University of Sydney in 2019, where my research explored topic modeling and community detection in complex systems. After completing my undergraduate studies, I worked as a software engineer at Huawei for two years, gaining practical experience in software development and large-scale data analysis. I then pursued a Master’s degree in Computer Science at UMass Amherst, where I conducted research in deep learning under the supervision of Prof. Evangelos Kalogerakis and Dmitry Petrov, focusing on applications in natural language processing (NLP), computer vision, and graphics.

News

Publications
  1. EMNLP
    Abhilasha Lodha*, Gayatri Belapurkar*, Saloni Chalkapurkar*, Yuanming Tao*, Reshmi Ghosh, Samyadeep Basu, Dmitrii Petrov, Soundararajan Srinivasan
    EMNLP Findings, 2023

  2. EPJ Data Science
    Charles C. Hyland, Yuanming Tao, Lamiae Azizi, Martin Gerlach, Tiago P. Peixoto, Eduardo G. Altmann
    EPJ Data Science, 2021

Projects
  1. NLP
    2023
    The key contribution of the project is its novel approach to the ECPE task through several methods, such as 1) framing the problem as a multi-span extractive question-answering problem, 2) evaluating the performance of LLMs on this task, and 3) recreating SOTA frameworks in Pytorch that were originally crafted in TensorFlow 1.x.

  2. NLP
    2019
    The thesis's main contribution is enhancing the network model with additional textual information, improving the classification of documents. Experiments on Wikipedia articles confirm that incorporating auxiliary information into topic models results in a better fit to the data and superior document classification.

  3. Comp. Math.
    2018
    The project's contribution includes the creation of a fully vectorized and memory-optimized codebase, enhancing the exploration of dynamic behaviors in a multi-mass pendulum chain and revealing its dependence on initial conditions and the number of masses involved.

Contact

  • Address: School of Mathematics and Statistics F07, University of Sydney NSW 2006, Australia
  • Office: Room 528, Carslaw Building
  • School Email: ytao5219 (at) uni (dot) sydney (dot) edu
  • Misc

    • I grew up in Ürümqi , Xinjiang.
    • For most Chinese people, my name is very easy to remember because it is spelled the same as Tao Yuanming, the famous poet from the Eastern Jin dynasty in ancient China.


    Last Update:

    This website is built with Hugo, utilizing the Minimal Light theme.