Lequan Yu

Assistant Professor@The University of Hong Kong·中国香港·OpenReview

研究方向

multimodal foundation models · biomedical data science · image reconstruction · CT image · multi-modal learning · graph learning · Computational pathology · medical image analysis · medical AI · AI in healthcare

Lequan Yu

CausalNovo: Advancing De Novo Peptide Sequencing via a Causality-Informed Framework

Horseshoe Splatting: Handling Structural Sparsity for Uncertainty-Aware Gaussian-Splatting Radiance Field Rendering

Bridging Radiology and Pathology Foundation Models via Concept-Based Multimodal Co-Adaptation

AssetFormer: Modular 3D Assets Generation with Autoregressive Transformer

MedMMV: A Controllable Multimodal Multi-Agent Framework for Reliable and Verifiable Clinical Reasoning

Latent-Space Reinforcement Learning for Image Segmentation

Fairness-Aware EHR Analysis via Structured Missing Pattern Modeling and Adversarial Low-Rank Adaptation

StyleAR: Customizing Multimodal Autoregressive Model for Style-Aligned Text-to-Image Generation

Amplifying Prominent Representations in Multimodal Learning via Variational Dirichlet Process

Variational Polya Tree

From Layers to States: A State Space Model Perspective to Deep Neural Network Layer Dynamics

From Token to Rhythm: A Multi-Scale Approach for ECG-Language Pretraining

Cross-Modal Alignment via Variational Copula Modelling

ToolBridge: An Open-Source Dataset to Equip LLMs with External Tool Capabilities

Cross-Modal Alignment via Variational Copula Modelling

Bayesian Nonparametric Survival Analysis via Deep Dirichlet Process

Deep ECG-Report Interaction Framework for Cross-Modal Representation Learning