← Search

Wanghu Chen

2 accepted papers

2026

DualTimesField: Rethinking Time Series as Continuous-Time Trends and Events

ICML 2026poster

Effective time series representation is critical for revealing temporal dynamics in many fields. However, existing approaches encounter fundamental limitations. Discrete-time representations struggle with irregular sampling and the tradeoff of fidelity and efficiency, while traditional implicit neur…

Cited by 0SourceScholar
2025

PhysDiff: A Physically-Guided Diffusion Model for Multivariate Time Series Anomaly Detection

NeurIPS 2025poster

Unsupervised anomaly detection of multivariate time series remains challenging in complex nonstationary dynamics, due to the high false-positive rates and limited interpretability. We propose PhysDiff, combining physics-guided decomposition with diffusion-based reconstruction, to address these issue…

Cited by 0SourceScholar