Summary

Ph.D. student in Machine Learning at Georgia Tech (Advisor: Shihao Yang). Research interests include foundation models, next‑generation sequence modeling for NLP/CV/time series, and linear attention mechanisms. Reviewer service: NeurIPS 2024 & 2025, ICLR 2025, ICML 2025, and IEEE Internet of Things Journal.

Education

• Ph.D. in Machine Learning

  Present
  Georgia Institute of Technology
• M.S. in Analytics

  2023-08
  Georgia Institute of Technology
• Bachelor in Logistics Engineering

  2020-06
  Tianjin University

Work Experience

• Data Scientist Intern

  2021-07 - 2022-07
  Tencent (Tencent Medical AI Lab - JARVIS Lab)

  Developed deep learning/statistical models and co-authored publications & patents for medical time-series applications.

  • Co-authored 2 papers on medical time series models and filed 6 patents as lead inventor.
  • Built models for medical insurance forecasting and epidemic monitoring.
  • Delivered tech solutions for medical facilities in 4 major cities, handling up to 10 million data points daily.
• Business Analyst Intern

  2021-01 - 2021-07
  Amazon (Amazon Private Brands)

  Applied ML and causal inference for cost analysis; built AWS-based dashboards to support sourcing decisions.

  • Improved sourcing performance via data-driven methods for cost analysis.
  • Developed AWS-based dashboard web apps that helped save over $100,000 per month.
• Research Assistant

  2020-09 - 2021-01
  Peking University (Guanghua School of Management)

  Contributed to projects in supply chain finance, risk hedging, and logistics network design.

  • Supported 10+ research projects and 4 papers.
  • Worked with stochastic programming, game theory, and structural equation models using CPLEX, Stata, AMOS, and Python.
• Strategic and Data Analytics Intern

  2020-07 - 2020-09
  Country Garden Group

  Strategy and analytics internship experience.

Skills

Publications

• ZeroS: Zero‑Sum Linear Attention for Efficient Transformer

  2025
  NeurIPS 2025 (Spotlight)
• Linear Transformers as VAR Models: Aligning Autoregressive Attention Mechanisms with Autoregressive Forecasting

  2025
  ICML 2025
• WAVE: Weighted Autoregressive Varying Gate for Time Series Forecasting

  2025
  ICML 2025
• In‑context Time Series Predictor

  2025
  ICLR 2025
• CATS: Enhancing Multivariate Time Series Forecasting by Constructing Auxiliary Time Series as Exogenous Variables

  2024
  ICML 2024
• ARM: Refining Multivariate Forecasting with Adaptive Temporal‑Contextual Learning

  2024
  ICLR 2024