Publications
* Equal contribution
ProtoMol: Enhancing Molecular Property Prediction via Prototype-Guided Multimodal Learning
Published in Briefings in Bioinformatics, 2025
A prototype-guided multimodal framework that aligns molecular graphs and textual descriptions via layer-wise cross-modal attention and a shared prototype space, improving property prediction and interpretability.
Multi-microgrids Peer to Peer Electricity-carbon joint Trading Method Based on Multi-agent Ethical Enhancement Technology
Published in IEEE Transactions on Industry Applications, 2025
We propose a P2P electricity–carbon joint trading method using multi-agent ethical enhancement with safe reinforcement learning and privacy protection to achieve low-carbon economic operation for multi-microgrids.
Heterogeneity-aware high-efficiency federated learning with hybrid synchronous-asynchronous splitting strategy
Published in Neural Networks, 2025
We propose a resource-aware adaptive model customization mechanism that adapts suitable models, mixes synchronous edge feature extraction with asynchronous global classifier updates, and uses baseline-prioritized weighted aggregation.
Clinical Expert Uncertainty Guided Generalized Label Smoothing for Medical Noisy Label Learning
Published in arXiv, 2025
We propose a clinical expert uncertainty-aware benchmark and label smoothing method to address label noise in medical image datasets created from clinical notes.
MvHo-IB: Multi-View High-Order Information Bottleneck for Brain Disorder Diagnosis
Published in MICCAI 2025, 2025
We propose a novel method to capture higher-order interactions while using information bottleneck to reduce redundancy between multi-view information for brain disorder diagnosis.