Xumeng Wen

Reinforcement Learning with Verifiable Rewards Implicitly Incentivizes Correct Reasoning in Base LLMs

Are Time-Series Foundation Models Deployment-Ready? A Systematic Study of Adversarial Robustness Across Domains

ProbTS: A Unified Toolkit to Probe Deep Time-series Forecasting

ElasTST: Towards Robust Varied-Horizon Forecasting with Elastic Time-Series Transformer

NewTime: Numerically Multi-Scaled Embedding for Large-Scale Time Series Pretraining

Towards Foundation Models for Learning on Tabular Data