Chanyoung Park

CompoDistill: Attention Distillation for Compositional Reasoning in Multimodal LLMs

Token-Efficient Item Representation via Images for LLM Recommender Systems

Beyond the Final Answer: Evaluating the Reasoning Trajectories of Tool-Augmented Agents

Personalized Visual Representation Alignment for Generative Multimodal Recommendation

SelfJudge: Faster Speculative Decoding via Self-Supervised Judge Verification

LLM-based Online Time Series Forecasting with Frequency-driven Pattern Recognition

IR-Agent: Expert-Inspired LLM Agents for Structure Elucidation from Infrared Spectra

scMoE: single-cell Multi-Modal Multi-Task Learning via Sparse Mixture-of-Experts

Adaptive Graph Rewiring to Mitigate Over-Squashing in Mesh-Based GNNs for Fluid Dynamics Simulations

Subgraph Federated Learning for Local Generalization

Training Robust Graph Neural Networks by Modeling Noise Dependencies

3D Interaction Geometric Pre-training for Molecular Relational Learning

Self-Supervised Diffusion Models for Electron-Aware Molecular Representation Learning

Disentangling Hyperedges through the Lens of Category Theory

Weakly Supervised Video Scene Graph Generation via Natural Language Supervision

Global Context-aware Representation Learning for Spatially Resolved Transcriptomics

Global Context-aware Representation Learning for Spatially Resolved Transcriptomics

Thickness-aware E(3)-Equivariant 3D Mesh Neural Networks

Naturality-Guided Hyperedge Disentanglement for Message Passing Hypergraph Neural Network

Toward Generalizability of Graph-based Imputation on Biomedical Tabular-based Missing Data

3D Interaction Geometric Pre-training for Molecular Relational Learning

Enhancing single-cell Multi-Modal Multi-Task Learning via Sparse Mixture-of-Experts

Oldie but Goodie: Re-illuminating Label Propagation on Graphs with Partially Observed Features