ICML 2026 accepted papers on Graph learning
206 of the 6,341 papers accepted at ICML 2026 (3.2%) are on Graph learning, compared with 5.2% at ICML 2025. They fall into 4 topics.
See ICML 2026 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.
Topics and papers
Graph self-supervised learning · 80 papers
- FLAG: Foundation model representation with Latent diffusion Alignment via Graph for spatial gene expression prediction (1 citation)
- Aitchison Embeddings for Learning Compositional Graph Representations
- Are Common Substructures Transferable? Riemannian Graph Foundation Model with Neural Vector Bundles
- Boundary Embedding Shaping with Adaptive Contrastive Learning for Graph Structural Disentanglement
- DANCE: Dynamic, Available, Neighbor-gated Condensation for Federated Text-Attributed Graphs
- Efficient Code Analysis via Graph Representation Learning-Guided Large Language Models
- Enhancing LLMs for Graph Tasks via Graph-aware LoRA Generation
- ERAlign: Energy-based Representation Alignment of GNNs and LLMs on Text-attributed Graphs
- Feature-Aware (Hyper)graph Generation via Next-Scale Prediction
- From Moments to Models: Graphon-Mixture Learning for Mixup and Contrastive Learning
- GCIB: Graph Contrastive Information Bottleneck for Multi-Behavior Recommendation
- Geometry-Guided Generative Representation for Functional Brain Graphs
and 68 more in the interactive map.
GNN architectures · 60 papers
- Backward Oversmoothing: why is it hard to train deep Graph Neural Networks? (1 citation)
- A Graphop Analysis of Graph Neural Networks on Sparse Graphs: Generalization and Universal Approximation
- A Unifying Relational Perspective on Expressive Lottery Tickets
- Adaptive Node Feature Selection for Graph Neural Networks
- Adaptive Recurrent Message Passing for Test Time Computing on Graphs
- Anchor-guided Hypergraph Condensation with Dual-level Discrimination
- Beyond ReLU: Bifurcation, Oversmoothing, and Topological Priors
- Can Computational Reducibility Lead to Transferable Models for Graph Combinatorial Optimization?
- DREAM: Dual-Standard Semantic Homogeneity with Dynamic Optimization for Graph Learning with Label Noise
- Efficient Prediction of SO(3)-Equivariant Hamiltonian Matrices via SO(2) Local Frames
- Fixed Aggregation Features Can Rival GNNs
- Foundations of Equivariant Deep Learning: Unifying Graph and Sheaf Neural Networks
and 48 more in the interactive map.
Graph algorithms and clustering · 39 papers
- A General Neural Backbone for Mixed-Integer Linear Optimization via Dual Attention
- Active Curriculum Refinement for Reinforcement Learning
- An Approximation Algorithm for Graph Label Selection
- Attributed Network Alignment: Statistical Limits and Efficient Algorithm
- Beyond Binary: Continuous State Optimization with Graph-Structured Objectives
- Bipartite Graph Attention-based Clustering for Large-scale scRNA-seq Data
- BlitzRank: Principled Zero-shot Ranking Agents with Tournament Graphs
- CauScale: Neural Causal Discovery at Scale
- Collapsed Effective Operators for Higher-order Structures
- Computationally-efficient Graph Modeling with Refined Graph Random Features
- D$^3$: Dynamic Directional Graph-Constrained Data Scheduling for LLM Training
- Estimating Correlation Clustering Cost in Node-Arrival Stream
and 27 more in the interactive map.
Knowledge and temporal graphs · 27 papers
- Generative Representation Learning on Hyper-relational Knowledge Graphs via Masked Discrete Diffusion (1 citation)
- A Tale of Two Graphs: Separating Knowledge Exploration from Outline Structure for Open-Ended Deep Research
- Adaptive Memory Retention in Dynamic Graphs
- Beyond Explicit Edges: Robust Reasoning over Noisy and Sparse Knowledge Graphs
- CausalRAG2: Hierarchical Causal Knowledge Graph Design for RAG
- Conformal Path Reasoning: Trustworthy Knowledge Graph Question Answering via Path-Level Calibration
- FusionCell: Cross-Attentive Fusion of Layout Geometry and Netlist Topology for Standard-Cell Performance Prediction
- GeoFlow: Geo-Aware Modeling of Inter-Area Relationships in Origin-Destination Flow Prediction and Generation
- Learning Long Range Spatio-Temporal Representations over Continuous Time Dynamic Graphs with State Space Models
- Mesh Based Simulations with Spatial and Temporal awareness
- PromptDyG: Test-Time Prompt Adaptation on Dynamic Graphs
- Rel-MOSS: Towards Imbalanced Relational Deep Learning on Relational Databases
and 15 more in the interactive map.