Scholar
Jianwei Wang
Google Scholar ID: fXt4M2kAAAAJ
University of New South Wales
Citations & Impact
Citations
76
H-index
4
i10-index
3
Publications
8
Co-authors
4
list available
Publications
Resume (English only)
Academic Achievements
- Preprints and under review:
- - Multivariate Time Series Imputation with Signal-Noise Disentangled Graph Propagation
- - EMFI: EM-Driven Node Feature Imputation with Context-Aware Information Enhancement
- - Ensemble-based Deep Multilayer Community Search
- - Empowering Tabular Data Preparation with Language Models: Why and How?
- - LLM-based HSE Compliance Assessment: Benchmark, Performance, and Advancements
- - AutoResearcher: Automating Knowledge-Grounded and Transparent Research Ideation with Multi-Agent Collaboration
- 2025 Publications:
- - On LLM-Enhanced Mixed-Type Data Imputation with High-Order Message Passing
- - ProbDiffFlow: An Efficient Learning-Free Framework for Probabilistic Single-Image Optical Flow Estimation
- - COMET: An Interactive Framework for Efficient and Effective Community Search via Active Learning
- - Account Risk Detection in Large-Scale Financial Graphs with Auxiliary Asset Prediction
- 2024 Publications:
- - Efficient Unsupervised Community Search with Pre-trained Graph Transformer
- - Missing Data Imputation with Uncertainty-Driven Network
- - Neural Attributed Community Search at Billion Scale
Research Experience
- Postdoctoral Research Fellow, University of New South Wales, Sydney, Australia, 2024.11 - Now
- Guest Lecturer (COMP9311, 2025T3, UNSW), 2025
- Casual Academic and Teaching Assistant, 2021 - 2024
Education
- PhD in Computer Science and Engineering, UNSW, 2025
- B.Eng. in Chu Kochen Honors College, Zhejiang University, 2021
Background
- Currently a postdoctoral research fellow at the University of New South Wales (UNSW), with research interests including: 1. Graph-enhanced foundation models for data preparation and management; 2. Multimodal data preparation and management; 3. Reinforced data training and quality optimization; 4. Explainable data preparation and interdisciplinary application, especially in quantum error correction, marketing strategy, biological drug design, financial security, and more.
Miscellany
- Personal interests not provided
