Qingmei Wang (王庆梅)   Ph.D. in Artificial Intelligence Gaoling School of Artificial Intelligence Renmin University of China Email: qingmeiwang@ruc.edu.cn Links: [Google Scholar] [Github]

Biography

I received my Ph.D. in Artificial Intelligence from the Gaoling School of Artificial Intelligence, Renmin University of China in 2026, advised by Prof. Hongteng Xu.

My research interests span temporal point processes, generative models, optimal transport, and reinforcement learning.

I am actively seeking postdoctoral or researcher positions and am available to start immediately. Please feel free to contact me if our interests align.

Research

I develop learning-based and diffusion-based methods for temporal point processes (TPPs). TPPs are probabilistic models of asynchronous event streams. My work focuses on robustness and generalization under incomplete data.

Real-world event data is rarely clean: events go unobserved, sequences are truncated, and distributions shift across cohorts and domains. My doctoral work builds TPPs that stay reliable under these conditions, following a single arc — from robust representation (AAAI 2023, Oral), to structured inference of event dependencies (AAAI 2025), to prediction on unseen sequences (AAAI 2026), to diffusion-based generative adaptation across distributions (manuscript in preparation).

I am now extending this line toward event foundation models and decision-making over learned event simulators. Applications span clinical, urban, and financial event streams.

Publications

• ST-TPP: Learning Semi-Transductive Temporal Point Processes with Gromov–Wasserstein Barycentric Regularization
  Qingmei Wang, Tianyu Huang, Yujie Long, Yuxin Wu, Junchi Yan, Hongteng Xu
  AAAI Conference on Artificial Intelligence (AAAI), 2026.
  Predicts on unseen event sequences by fusing inductive and transductive information via optimal-transport barycenters — effective in cold-start settings.
  [Paper]

• A Plug-and-Play Bregman ADMM Module for Inferring Event Branches in Temporal Point Processes
  Qingmei Wang, Yuxin Wu, Yujie Long, Jing Huang, Fengyuan Ran, Bing Su, Hongteng Xu
  AAAI Conference on Artificial Intelligence (AAAI), 2025.
  A drop-in module that infers structured event dependencies under sparse and low-rank constraints, improving existing TPP models.
  [Paper] [Code]

• Hierarchical Contrastive Learning for Temporal Point Processes
  Qingmei Wang, Minjie Cheng, Shen Yuan, Hongteng Xu
  AAAI Conference on Artificial Intelligence (AAAI), 2023.
  Oral Presentation, AAAI 2023
  Combines event-level and sequence-level contrastive learning, reducing sampling complexity from O(N²) to O(N).
  [Paper]

Honors & Awards

• Oral Presentation, AAAI 2023
• Merit Student, Renmin University of China, 2024
• Academic Excellence Scholarship, Renmin University of China (multiple years)
• Outstanding Undergraduate Thesis (university level)

Academic Service

Conference Reviewer: ICML (2023–2024), NeurIPS (2023–2024), ICLR (2024), KDD (2024–2026), AAAI (2026)
Teaching Assistant: Foundations of Machine Learning (undergraduate); Massive Data Mining (graduate)