Defu Cao

PINFDiT: Energy-Based Physics-Informed Diffusion Transformers for General-purpose Time Series Tasks

Adaptive Collaboration with Humans: Metacognitive Policy Optimization for Multi-Agent LLMs with Continual Learning

ClimateLLM: Efficient Weather Forecasting via Frequency-Aware Large Language Models

Learning What Matters: Toward Causally-Grounded Foundations for Contextual Time-Series Forecasting

Conversational Time Series Foundation Models: Towards Explainable and Effective Forecasting

Selecting Out-of-Distribution Detector for Multiple Modalities

When LLM Meets Time Series: A Real-World Benchmark for Explicit and Implicit Multi-Step Reasoning

When Splitting Makes Stronger: A Theoretical and Empirical Analysis of Divide-and-Conquer Prompting in LLMs

Neuron-based Multifractal Analysis of Neuron Interaction Dynamics in Large Models

Orthogonalized Estimation of Difference of Q-functions

TimeDiT: General-purpose Diffusion Transformers for Time Series Foundation Model

KambaAD: Enhancing State Space Models with Kolmogorov–Arnold for time series Anomaly Detection

An Examination on the Effectiveness of Divide-and-Conquer Prompting in Large Language Models

Beyond Forecasting: Compositional Time Series Reasoning for End-to-End Task Execution