Sheng Zhou

RRVF: Visual Reinforcement Learning with Reasoning, Rendering, and Visual Feedback

Making Classic GNNs Strong Baselines Across Varying Homophily: A Smoothness–Generalization Perspective

Understanding and Enhancing Message Passing on Heterophilic Graphs via Compatibility Matrix

DocGenome: A Large Benchmark for Multi-Modal Language Models in Real-World Academic Document Understanding

Towards a Unified Framework of Clustering-based Anomaly Detection

Towards Dynamic Graph Neural Networks with Provably High-Order Expressive Power

PruningBench: A Comprehensive Benchmark of Structural Pruning

Towards a Unified Framework of Clustering-based Anomaly Detection

Exploring and Unleashing the Power of Message Passing on Heterophilous Graphs