Publications
† Equal contribution * Co-corresponding author
Conference
-
NeurIPSDisentangling Hyperedges through the lens of Category TheoryThe Thirty-Ninth Annual Conference on Neural Information Processing Systems (NeurIPS), 2025AbstractPDF Code
Despite the promising results of disentangled representation learning in discovering latent patterns in graph-structured data, few studies have explored disentanglement for hypergraph-structured data. Integrating hyperedge disentanglement into hypergraph neural networks enables models to leverage hidden hyperedge semantics, such as unannotated relations between nodes, that are associated with labels. This paper presents an analysis of hyperedge disentanglement from a category-theoretical perspective and proposes a novel criterion for disentanglement derived from the naturality condition. Our proof-of-concept model experimentally showed the potential of the proposed criterion by successfully capturing functional relations of genes (nodes) in genetic pathways (hyperedges).
-
ICMLThickness-aware E(3)-Equivariant 3D Mesh Neural NetworksForty-Second International Conference on Machine Learning (ICML), 2025AbstractPDF Slides
Mesh-based 3D static analysis methods have recently emerged as efficient alternatives to traditional computational numerical solvers, significantly reducing computational costs and runtime for various physics-based analyses. However, these methods primarily focus on surface topology and geometry, often overlooking the inherent thickness of real-world 3D objects, which exhibits high correlations and similar behavior between opposing surfaces. This limitation arises from the disconnected nature of these surfaces and the absence of internal edge connections within the mesh. In this work, we propose a novel framework, the Thickness-aware E(3)-Equivariant 3D Mesh Neural Network (T-EMNN), that effectively integrates the thickness of 3D objects while maintaining the computational efficiency of surface meshes. Additionally, we introduce data-driven coordinates that encode spatial information while preserving E(3)-equivariance or invariance properties, ensuring consistent and robust analysis. Evaluations on a real-world industrial dataset demonstrate the superior performance of T-EMNN in accurately predicting node-level 3D deformations, effectively capturing thickness effects while maintaining computational efficiency.
-
ICLRSubgraph Federated Learning for Local GeneralizationThe Thirteenth International Conference on Learning Representations (ICLR) (Oral, top 1.8%), 2025AbstractPDF Code Slides
Federated Learning (FL) on graphs enables collaborative model training to enhance performance without compromising the privacy of each client. However, existing methods often overlook the mutable nature of graph data, which frequently introduces new nodes and leads to shifts in label distribution. Since they focus solely on performing well on each client’s local data, they are prone to overfitting to their local distributions (i.e., local overfitting), which hinders their ability to generalize to unseen data with diverse label distributions. In contrast, our proposed method, FedLoG, effectively tackles this issue by mitigating local overfitting. Our model generates global synthetic data by condensing the reliable information from each class representation and its structural information across clients. Using these synthetic data as a training set, we alleviate the local overfitting problem by adaptively generalizing the absent knowledge within each local dataset. This enhances the generalization capabilities of local models, enabling them to handle unseen data effectively. Our model outperforms baselines in our proposed experimental settings, which are designed to measure generalization power to unseen data in practical scenarios.
-
KDDSelf-Explainable Temporal Graph Networks based on Graph Information BottleneckIn ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2024AbstractPDF Code Slides
Temporal Graph Neural Networks (TGNN) have the ability to capture both the graph topology and dynamic dependencies of interactions within a graph over time. There has been a growing need to explain the predictions of TGNN models due to the difficulty in identifying how past events influence their predictions. Since the explanation model for a static graph cannot be readily applied to temporal graphs due to its inability to capture temporal dependencies, recent studies proposed explanation models for temporal graphs. However, existing explanation models for temporal graphs rely on post-hoc explanations, requiring separate models for prediction and explanation, which is limited in two aspects: efficiency and accuracy of explanation. In this work, we propose a novel built-in explanation framework for temporal graphs, called Self-Explainable Temporal Graph Networks based on Graph Information Bottleneck (TGIB). TGIB provides explanations for event occurrences by introducing stochasticity in each temporal event based on the Information Bottleneck theory. Experimental results demonstrate the superiority of TGIB in terms of both the link prediction performance and explainability compared to state-of-the-art methods. This is the first work that simultaneously performs prediction and explanation for temporal graphs in an end-to-end manner.
-
ICMLUnsupervised Episode Generation for Graph Meta-learningInternational Conference on Machine Learning (ICML), 2024AbstractPDF Code Slides
We propose Unsupervised Episode Generation method called Neighbors as Queries (NaQ) to solve the Few-Shot Node-Classification (FSNC) task by unsupervised Graph Meta-learning. Doing so enables full utilization of the information of all nodes in a graph, which is not possible in current supervised meta-learning methods for FSNC due to the label-scarcity problem. In addition, unlike unsupervised Graph Contrastive Learning (GCL) methods that overlook the downstream task to be solved at the training phase resulting in vulnerability to class imbalance of a graph, we adopt the episodic learning framework that allows the model to be aware of the downstream task format, i.e., FSNC. The proposed NaQ is a simple but effective unsupervised episode generation method that randomly samples nodes from a graph to make a support set, followed by similarity-based sampling of nodes to make the corresponding query set. Since NaQ is model-agnostic, any existing supervised graph meta-learning methods can be trained in an unsupervised manner, while not sacrificing much of their performance or sometimes even improving them. Extensive experimental results demonstrate the effectiveness of our proposed unsupervised episode generation method for graph meta-learning towards the FSNC task.
-
WWWDSLR: Diversity Enhancement and Structure Learning for Rehearsal-based Graph Continual LearningThe Web Conference (WWW) (Oral), 2024AbstractPDF Code Slides
We investigate the replay buffer in rehearsal-based approaches for graph continual learning (GCL) methods. Existing rehearsal-based GCL methods select the most representative nodes for each class and store them in a replay buffer for later use in training subsequent tasks. However, we discovered that considering only the class representativeness of each replayed node makes the replayed nodes to be concentrated around the center of each class, incurring a potential risk of overfitting to nodes residing in those regions, which aggravates catastrophic forgetting. Moreover, as the rehearsal-based approach heavily relies on a few replayed nodes to retain knowledge obtained from previous tasks, involving the replayed nodes that have irrelevant neighbors in the model training may have a significant detrimental impact on model performance. In this paper, we propose a GCL model named DSLR, specifically, we devise a coverage-based diversity (CD) approach to consider both the class representativeness and the diversity within each class of the replayed nodes. Moreover, we adopt graph structure learning (GSL) to ensure that the replayed nodes are connected to truly informative neighbors. Extensive experimental results demonstrate the effectiveness and efficiency of DSLR.
-
NeurIPSInterpretable Prototype-based Graph Information BottleneckIn Conference on Neural Information Processing Systems (NeurIPS) (Gold Prize at the 2023 Samsung Humantech Paper Award), 2023AbstractPDF Code
The success of Graph Neural Networks (GNNs) has led to a need for understanding their decision-making process and providing explanations for their predictions, which has given rise to explainable AI (XAI) that offers transparent explanations for black-box models. Recently, the use of prototypes has successfully improved the explainability of models by learning prototypes to imply training graphs that affect the prediction. However, these approaches tend to provide prototypes with excessive information from the entire graph, leading to the exclusion of key substructures or the inclusion of irrelevant substructures, which can limit both the interpretability and the performance of the model in downstream tasks. In this work, we propose a novel framework of explainable GNNs, called interpretable Prototype-based Graph Information Bottleneck (PGIB) that incorporates prototype learning within the information bottleneck framework to provide prototypes with the key subgraph from the input graph that is important for the model prediction. This is the first work that incorporates prototype learning into the process of identifying the key subgraphs that have a critical impact on the prediction performance. Extensive experiments, including qualitative analysis, demonstrate that PGIB outperforms state-of-the-art methods in terms of both prediction performance and explainability.
-
NeurIPSDensity of States Prediction of Crystalline Materials via Prompt-guided Multi-Modal TransformerIn Conference on Neural Information Processing Systems (NeurIPS), 2023AbstractPDF Code
The density of states (DOS) is a spectral property of crystalline materials, which provides fundamental insights into various characteristics of the materials. While previous works mainly focus on obtaining high-quality representations of crystalline materials for DOS prediction, we focus on predicting the DOS from the obtained representations by reflecting the nature of DOS: DOS determines the general distribution of states as a function of energy. That is, DOS is not solely determined by the crystalline material but also by the energy levels, which has been neglected in previous works. In this paper, we propose to integrate heterogeneous information obtained from the crystalline materials and the energies via a multi-modal transformer, thereby modeling the complex relationships between the atoms in the crystalline materials and various energy levels for DOS prediction. Moreover, we propose to utilize prompts to guide the model to learn the crystal structural system-specific interactions between crystalline materials and energies. Extensive experiments on two types of DOS, i.e., Phonon DOS and Electron DOS, with various real-world scenarios demonstrate the superiority of DOSTransformer.
-
KDDTask-Equivariant Graph Few-shot LearningIn ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2023AbstractPDF Code Slides
Although Graph Neural Networks (GNNs) have been successful in node classification tasks, their performance heavily relies on the availability of a sufficient number of labeled nodes per class. In real-world situations, not all classes have many labeled nodes and there may be instances where the model needs to classify new classes, making manual labeling difficult. To solve this problem, it is important for GNNs to be able to classify nodes with a limited number of labeled nodes, known as few-shot node classification. Previous episodic meta-learning based methods have demonstrated success in few-shot node classification, but our findings suggest that optimal performance can only be achieved with a substantial amount of diverse training meta-tasks. To address this challenge of meta-learning based few-shot learning (FSL), we propose a new approach, the Task-Equivariant Graph few-shot learning (TEG) framework. Our TEG framework enables the model to learn transferable task-adaptation strategies using a limited number of training meta-tasks, allowing it to acquire meta-knowledge for a wide range of meta-tasks. By incorporating equivariant neural networks, TEG can utilize their strong generalization abilities to learn highly adaptable task-specific strategies. As a result, TEG achieves state-of-the-art performance with limited training meta-tasks. Our experiments on various benchmark datasets demonstrate TEG’s superiority in terms of accuracy and generalization ability, even when using minimal meta-training data, highlighting the effectiveness of our proposed approach in addressing the challenges of meta-learning based few-shot node classification.
-
ICMLConditional Graph Information Bottleneck for Molecular Relational LearningIn International Conference on Machine Learning (ICML), 2023AbstractPDF Code Slides
Molecular relational learning, whose goal is to learn the interaction behavior between molecular pairs, got a surge of interest in molecular sciences due to its wide range of applications. Recently, graph neural networks have shown great success in molecular relational learning by modeling a molecule as a graph structure, and considering atom-level interactions between two molecules. Despite their success, existing molecular relational learning methods tend to overlook the nature of chemistry, i.e., a chemical compound is composed of multiple substructures such as functional groups that cause distinctive chemical reactions. In this work, we propose a novel relational learning framework, called CGIB, that predicts the interaction behavior between a pair of graphs by detecting core subgraphs therein. The main idea is, given a pair of graphs, to find a subgraph from a graph that contains the minimal sufficient information regarding the task at hand conditioned on the paired graph based on the principle of conditional graph information bottleneck. We argue that our proposed method mimics the nature of chemical reactions, i.e., the core substructure of a molecule varies depending on which other molecule it interacts with. Extensive experiments on various tasks with real-world datasets demonstrate the superiority of CGIB over state-of-the-art baselines.
Journal
-
Water ResearchPhysics-Embedded Graph Neural Operator for Interaction-Controlled Colloidal AggregationWater Research, 2026AbstractCode
Colloidal aggregation in natural and engineered waters is a complex function of both particle and groundwater electrochemical conditions, yet predicting aggregation behavior across diffusion-limited and reaction-limited regimes remains challenging due to the nonlinear dependence of collision efficiency on ionic strength and surface potential. This study presents a graph neural operator surrogate model that captures particle aggregation dynamics by embedding transport physics directly into the network architecture. Particle size classes are represented as graph nodes with Brownian collision kernels encoded in edge features, while attention mechanisms conditioned on ionic strength and zeta potential learn collision efficiency variations governed by extended DLVO interactions. The proposed model predicts aggregation kinetics across a range of electrochemical conditions, achieving R2 > 0.99 for both held-out parameter combinations and temporal extrapolation, and outperforming baseline models that do not embed physics in the architecture, as well as loss-based, physics-regularized neural networks. Validation against experimental measurements from bacteriophage, polystyrene, and cerium oxide systems confirms reproduction of aggregation regime transitions and kinetic saturation. Attention analysis reveals physically consistent reorganization from multiple information pathways under strong electrostatic repulsion to integrated processing under weak repulsion. This framework enables rapid, physically grounded exploration of colloidal aggregation relevant to water quality prediction and treatment optimization.
-
BioinformaticsDeep Single-cell RNA-seq Data Clustering with Graph Prototypical Contrastive LearningBioinformatics (SCI), 2023AbstractPDF Code
Single-cell RNA sequencing (scRNA-seq) enables researchers to study cellular heterogeneity by measuring transcriptome-wide gene expression at single cell level. To this end, identifying subgroups of cells with clustering techniques becomes an important task for downstream analysis. However, challenges on the scRNA-seq data such as pervasive dropout phenomena and high dimensionality hinder obtaining robust clustering outputs. Although many existing works are proposed to alleviate these problems, we argue that they fall short of fully leveraging the relational information inherent in the data, and most of them only adopt reconstruction-based losses that highly depend on the quality of features. In this paper, we propose a graph-based prototypical contrastive learning method, named scGPCL. Specifically, given a cell-gene bipartite graph that captures the natural relationship inherent in the scRNA-seq data, scGPCL encodes the cell representations based on Graph Neural Networks (GNNs), and utilizes prototypical contrastive learning scheme to learn cell representations by pushing apart semantically disimillar pairs and pulling together similar ones. Through extensive experiments on both simulated and real scRNA-seq data, we demonstrate that scGPCL not only obtains robust cell clustering outputs, but also handles the large-scale scRNA-seq data.
Workshop
-
ICLR WorkshopEqGINO: Equivariant Geometry-Informed Fourier Neural Operators for 3D PDEsICLR 2026 Workshop on AI and Partial Differential Equations (AI&PDE) (Oral), 2026AbstractPDF Code
Deep learning surrogates for 3D Partial Differential Equations (PDEs) often fail to generalize across geometric transformations because they depend heavily on specific coordinate systems. While equivariant networks offer a solution, they typically rely on local operations in the spatial domain, making the global receptive field—essential for PDE dynamics—computationally expensive. Conversely, Fourier Neural Operators (FNOs) efficiently capture global interactions, yet establishing 3D equivariance within them remains impractical due to the prohibitive cost of spectral group convolutions. To bridge this gap, we introduce EqGINO, a geometrically robust framework that enforces isotropy in the spectral domain. By design, EqGINO guarantees exact equivariance to the discrete symmetries inherent to the discretized computational domain. Beyond this discrete guarantee, our structural prior enables effective generalization to arbitrary continuous orientations even with a limited number of SE(3)-transformed training samples. Consequently, our method robustly models coordinate-invariant physical laws on complex irregular 3D geometries.
-
ICLR WorkshopCapturing Functional Context of Genetic Pathways through Hyperedge DisentanglementICLR 2025 Workshop on Machine Learning for Genomics Explorations (MLGenX), 2025
-
KDD WorkshopSubgraph Federated Learning for Local GeneralizationKDD Workshop on Federated Learning for Data Mining and Graph Analytics (FedKDD 2024) (Best Paper Award), 2024Abstract
Federated Learning (FL) on graphs enables collaborative model training to enhance performance without compromising the privacy of each client. However, existing methods often overlook the mutable nature of graph data, which frequently introduces new nodes and leads to shifts in label distribution. Since they focus solely on performing well on each client’s local data, they are prone to overfitting to their local distributions (i.e., local overfitting), which hinders their ability to generalize to unseen data with diverse label distributions. In contrast, our proposed method, FedLoG, effectively tackles this issue by mitigating local overfitting. Our model generates global synthetic data by condensing the reliable information from each class representation and its structural information across clients. Using these synthetic data as a training set, we alleviate the local overfitting problem by adaptively generalizing the absent knowledge within each local dataset. This enhances the generalization capabilities of local models, enabling them to handle unseen data effectively. Our model outperforms baselines in our proposed experimental settings, which are designed to measure generalization power to unseen data in practical scenarios.
-
KDD WorkshopInterpretable Graph Model with Prototype-Based Graph Information BottleneckKDD 2024 Workshop on Human-Interpretable AI (HI-AI 2024) (Best Paper Award), 2024
-
ICML WorkshopDeep Single-cell RNA-seq data Clustering with Graph Prototypical Contrastive LearningICML Workshop on Computational Biology (WCB 2023), 2023AbstractPDF Code
Single-cell RNA sequencing (scRNA-seq) enables researchers to study cellular heterogeneity by measuring transcriptome-wide gene expression at single cell level. To this end, identifying subgroups of cells with clustering techniques becomes an important task for downstream analysis. However, challenges on the scRNA-seq data such as pervasive dropout phenomena and high dimensionality hinder obtaining robust clustering outputs. Although many existing works are proposed to alleviate these problems, we argue that they fall short of fully leveraging the relational information inherent in the data, and most of them only adopt reconstruction-based losses that highly depend on the quality of features. In this paper, we propose a graph-based prototypical contrastive learning method, named scGPCL. Specifically, given a cell-gene bipartite graph that captures the natural relationship inherent in the scRNA-seq data, scGPCL encodes the cell representations based on Graph Neural Networks (GNNs), and utilizes prototypical contrastive learning scheme to learn cell representations by pushing apart semantically disimillar pairs and pulling together similar ones. Through extensive experiments on both simulated and real scRNA-seq data, we demonstrate that scGPCL not only obtains robust cell clustering outputs, but also handles the large-scale scRNA-seq data.
-
ICLR WorkshopPredicting Density of States via Multi-modal TransformerIn ICLR Workshop on Machine Learning for Materials (ML4Materials), 2023AbstractPDF Code
The density of states (DOS) is a spectral property of materials, which provides fundamental insights on various characteristics of materials. In this paper, we propose a model to predict the DOS by reflecting the nature of DOS: DOS determines the general distribution of states as a function of energy. Specifically, we integrate the heterogeneous information obtained from the crystal structure and the energies via multi-modal transformer, thereby modeling the complex relationships between the atoms in the crystal structure, and various energy levels. Extensive experiments on two types of DOS, i.e., Phonon DOS and Electron DOS, with various real-world scenarios demonstrate the superiority of DOSTransformer.