Xingjian Wu

TFB-2: benchmarking and automated ensemble for time series forecasting

FLAME: Flow Enhanced Legendre Memory Models for General Time Series Forecasting

In this work, we introduce FLAME, a family of extremely lightweight and capable Time Series Foundation Models, which support versatile forecasting tasks via generative probabilistic modeling, while ensuring both efficiency and robustness. FLAME utilizes the Legendre Memory for strong generalization capabilities. By ada...

Aurora-X: Built for Extreme Time Series Forecasting

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...

ST-EVO: Towards Generative Spatio-Temporal Evolution of Multi-Agent Communication Topologies

ST-EVO is proposed, which supports dialogue-wise communication scheduling with a compact yet powerful flow-matching based Scheduler to make precise Spatio-Temporal scheduling, and can also perceive the uncertainty of MAS, and possesses self-feedback ability to learn from accumulated experience.

ClawTrack: Towards Trace-Level Evaluation and Improvement of Real-World Autonomous Agents

ClawTrack is presented, a dual-assessment benchmark that simultaneously measures what an agent achieves (Task Score) and how it achieves it (Process Score) and finds that process scores effectively attribute success and failure to specific reasoning dimensions, filtering lucky passes invisible to outcome-only evaluatio...

Contrastive Reinforced Policy Optimization via Privileged Self-Distillation

Contrastive Reinforced Policy Optimization (CRPO) is introduced, which reformulates agentic OPSD from a contrastive learning perspective, and conducts group-wise contrast to preserve reliable, fine-grained optimization signals.