Research map / Graph learning / GNN architectures
Explore In-Context Message Passing Operator for Graph Neural Networks in A Mean Field Game
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Frames GNN feature learning as a mean-field game in which each node is an agent, and proposes an in-context message passing operator to replace the identical message passing used by current GNNs.
Problem
Current GNNs apply identical message passing operators to all nodes, which the authors argue is a limitation when node feature learning is viewed as a mean-field game among topologically connected agents.
Approach
Conceptualizes GNN learning as a mean-field game where each node is an agent interacting with its neighbors, and introduces an in-context message passing operator so that message passing is adapted per node rather than identical across nodes.
Experimental setup
Builds on: message passing neural networks mean-field game theory
Closest papers by meaning
- Learning Mean Field Games on Sparse Graphs: A Hybrid Graphex Approach (ICLR 2024 · similarity 0.89)
- Towards Dynamic Message Passing on Graphs (NeurIPS 2024 · similarity 0.88)
- Adaptive Message Passing: A General Framework to Mitigate Oversmoothing, Oversquashing, and Underreaching (ICML 2025 · similarity 0.88)
- Learning Latent Structures in Network Games via Data-Dependent Gated-Prior Graph Variational Autoencoders (ICML 2024 · similarity 0.88)
- Accurate and Scalable Graph Neural Networks via Message Invariance (ICLR 2025 · similarity 0.87)
- What do Graph Neural Networks learn? Insights from Tropical Geometry (NeurIPS 2024 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.