Jieming Shi

FACT: Fine-grained Across-variable Convolution for Multivariate Time Series Forecasting

Prior-free Tabular Test-time Adaptation

TimesMR: Unlocking the Potential of MLP and RNNs for Multivariate Time Series Forecasting

Token-Efficient Long-Term Interest Sketching and Internalized Reasoning for LLM-based Recommendation

Strengthen Out-of-Distribution Detection with Uncertainty-Driven Adaptively Rectified Backpropagation

TINED: GNNs-to-MLPs by Teacher Injection and Dirichlet Energy Distillation

SteerConf: Steering LLMs for Confidence Elicitation