Deep generative models have achieved remarkable success in time series generation, yet they largely neglect interventional and counterfactual inference. Existing causal generative methods typically rely on discrete environment prototypes, which fail to capture the continuous nature of real-world confounding. To address...
Changchen Song, Peng Chen, Yang Shu· Proceedings of the 32nd ACM...· 0 citations
A novel pattern-guided mixture-of-experts that expands model capacity through sparse activation and uses shallow patch similarities to constrain deep-layer routing, guiding expert specialization across heterogeneous time series and an implicit quantile network head that predicts arbitrary quantiles to characterize pred...
A unified benchmark for multimodal time series forecasting with heterogeneous context, MUSE-Bench enables systematic evaluation of how forecasting models utilize context and provides a foundation for future multimodal forecasting research.