v269: Proceedings of Learning on Graphs 2025

2026-06-04 · Source: Proceedings of Machine Learning Research · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics · Depth: Expert, medium

Summary

Volume 269 of the Proceedings of the Third Learning on Graphs Conference, held virtually from November 26-29, 2024, presents 47 research papers covering diverse advancements in graph learning. Key oral presentations include revisiting graph homophily measures, UnRavL for neuro-symbolic graph pattern queries, and a general recipe for combinatorial optimization with Multi-Filter GNNs. Poster presentations explore understanding GNN representations, task-specific graph subsampling via the Ising Model, and reinforcement learning for decentralized graph path search. Further contributions address temporal graph networks, hypergraph neural networks, and applications in areas like crystal property prediction with CrysAtom, power grid forecasting with NP-NDS, and dynamic representations of global crises using temporal knowledge graphs. The volume also features work on graph neural network expressivity, generalization error, and efficiency improvements.

Key takeaway

For AI Scientists and Machine Learning Engineers focused on graph-based models, this conference volume offers a comprehensive snapshot of current research trends. You should review the oral and poster presentations to identify emerging techniques in GNN architectures, knowledge graph reasoning, and temporal graph analysis. Consider exploring specific papers like UnRavL for neuro-symbolic queries or CrysAtom for material science applications to inform your project directions and potential collaborations.

Key insights

The conference highlights diverse advancements in graph learning, spanning theoretical foundations, architectural innovations, and real-world applications.

Topics

• Graph Neural Networks
• Knowledge Graphs
• Temporal Graph Learning
• Graph Optimization
• Hypergraph Neural Networks
• Graph Learning Theory

Code references

• WenkelF/copt
• rnartallo/decomposingflows
• giannisnik/gnn-representations
• mariabankestad/IsingOnGraphs
• flxclxc/rl-graph-search

Best for: AI Scientist, Machine Learning Engineer, Research Scientist

Related on AIssential

• LLM Friendly Graphs for AI Memory — From Triples to Facts — Larger, context-rich labels and subgraph representations enhance LLM comprehension of complex facts in knowledge graphs.
• A Gentle Introduction to Graph Neural Networks — GNNs leverage graph structures to model relationships, enabling predictions across diverse data types and tasks.
• Neural-Symbolic Logic Query Answering in Non-Euclidean Space — Hyperbolic GNNs with learned curvature effectively capture hierarchical logic query structures for knowledge graph reasoning.
• Predicting new research directions in materials science using large language models and concept graphs — LLMs and concept graphs can predict novel materials science research directions by identifying previously uncombined concepts.
• IEEE Transactions on Neural Networks and Learning Systems, Volume 37, Issue 4, April 2026 — The volume showcases diverse neural network and learning system advancements across various applications and theoretical foundations.
• SkillDAG: Self-Evolving Typed Skill Graphs for LLM Skill Selection at Scale — Dynamic, graph-based structural retrieval significantly enhances LLM agent skill selection and performance over static methods.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Proceedings of Machine Learning Research.