Daily arXiv Papers

Beyond Message Passing: A Symbolic Alternative for Expressive and Interpretable Graph Learning

GNN Graph Learning

SymGraph introduces a symbolic framework to go beyond traditional message passing in graph learning. By replacing standard GNN backbones with symbolic reasoning, it aims to overcome the 1-WL expressivity barrier and improve interpretability. The paper demonstrates enhanced expressive power and provides routes to more transparent models.

Spatio-temporal dual-stage hypergraph MARL for human-centric multimodal corridor traffic signal control

Graph Learning

STDSH-MARL proposes a scalable MARL framework for human-centric corridor traffic signal control. It uses a Spatio-Temporal Dual-Stage Hypergraph to capture dependencies among multimodal travelers under centralized training and decentralized execution. Experiments show improved multimodal throughput and reduced delays compared to baselines.

AdvSynGNN: Structure-Adaptive Graph Neural Nets via Adversarial Synthesis and Self-Corrective Propagation

GNN Graph Learning

AdvSynGNN presents a resilient GNN architecture that uses adversarial synthesis and self-corrective propagation to cope with structural noise and heterophily. It combines multi-resolution structure synthesis with contrastive objectives to produce geometry-aware initializations and a transformer backbone that adaptively handles heterophily by modulating attention. The result is more robust node representations across diverse graph topologies.

A Locality Radius Framework for Understanding Relational Inductive Bias in Database Learning

GNN Graph Learning

Note: The user-provided entry contains the title only without an English abstract. Please provide the abstract for a proper summary.

From Subtle to Significant: Prompt-Driven Self-Improving Optimization in Test-Time Graph OOD Detection

Graph Learning

Note: The user-provided entry contains the title only without an English abstract. Please provide the abstract for a proper summary.

LLM-WikiRace: Benchmarking Long-term Planning and Reasoning over Real-World Knowledge Graphs

Knowledge Graph LLM × Graph

LLM-WikiRace benchmarks long-term planning and reasoning over real-world knowledge graphs. In this task, models must navigate Wikipedia hyperlinks step by step to reach a target page, requiring look-ahead planning and understanding concept relationships. The study evaluates diverse LLMs and highlights planning capabilities and limits.

Semi-Supervised Learning on Graphs using Graph Neural Networks

GNN Graph Learning

Note: The user-provided entry contains the title only without an English abstract. Please provide the abstract for a proper summary.

GGBall: Graph Generative Model on Poincar\'e Ball

Graph Learning

GGBall proposes a hyperbolic graph generator on the Poincaré ball by integrating a Hyperbolic VQ-VAE with a Riemannian flow matching prior defined via closed-form geodesics. This design enables geometry-aware priors to model hierarchical structures in graphs.

Temporal Graph Pattern Machine

Graph Learning

Temporal Graph Pattern Machine (TGPM) explicitly models evolving temporal patterns in graphs to reveal transferable temporal evolution mechanisms. It moves beyond task-specific heuristics toward discovering underlying dynamics that generalize across settings.

Diffusion-Guided Pretraining for Brain Graph Foundation Models

Graph Learning

Note: The user-provided entry contains the title only without an English abstract. Please provide the abstract for a proper summary.

Retrieval Augmented (Knowledge Graph), and Large Language Model-Driven Design Structure Matrix (DSM) Generation of Cyber-Physical Systems

Knowledge Graph LLM × Graph

Note: The user-provided entry contains the title only without an English abstract. Please provide the abstract for a proper summary.

Instructor-Aligned Knowledge Graphs for Personalized Learning

Knowledge Graph

Note: The user-provided entry contains the title only without an English abstract. Please provide the abstract for a proper summary.

Visual Model Checking: Graph-Based Inference of Visual Routines for Image Retrieval

Graph Learning

Note: The user-provided entry contains the title only without an English abstract. Please provide the abstract for a proper summary.

Enhancing Large Language Models (LLMs) for Telecom using Dynamic Knowledge Graphs and Explainable Retrieval-Augmented Generation

Knowledge Graph LLM × Graph Graph Learning

Note: The user-provided entry contains the title only without an English abstract. Please provide the abstract for a proper summary.

GPU-Accelerated Algorithms for Graph Vector Search: Taxonomy, Empirical Study, and Research Directions

Graph Theory

Note: The user-provided entry contains the title only without an English abstract. Please provide the abstract for a proper summary.

Probability-Invariant Random Walk Learning on Gyral Folding-Based Cortical Similarity Networks for Alzheimer's and Lewy Body Dementia Diagnosis

Graph Learning

Note: The user-provided entry contains the title only without an English abstract. Please provide the abstract for a proper summary.

Daily arXiv Papers

Graph Neural Networks · Graph Learning · LLM × Graph

Conferences

Archives

Beyond Message Passing: A Symbolic Alternative for Expressive and Interpretable Graph Learning

Spatio-temporal dual-stage hypergraph MARL for human-centric multimodal corridor traffic signal control

AdvSynGNN: Structure-Adaptive Graph Neural Nets via Adversarial Synthesis and Self-Corrective Propagation

A Locality Radius Framework for Understanding Relational Inductive Bias in Database Learning

From Subtle to Significant: Prompt-Driven Self-Improving Optimization in Test-Time Graph OOD Detection

LLM-WikiRace: Benchmarking Long-term Planning and Reasoning over Real-World Knowledge Graphs

Semi-Supervised Learning on Graphs using Graph Neural Networks

GGBall: Graph Generative Model on Poincar\'e Ball

Temporal Graph Pattern Machine

Diffusion-Guided Pretraining for Brain Graph Foundation Models

Retrieval Augmented (Knowledge Graph), and Large Language Model-Driven Design Structure Matrix (DSM) Generation of Cyber-Physical Systems

Instructor-Aligned Knowledge Graphs for Personalized Learning

Visual Model Checking: Graph-Based Inference of Visual Routines for Image Retrieval

Enhancing Large Language Models (LLMs) for Telecom using Dynamic Knowledge Graphs and Explainable Retrieval-Augmented Generation

GPU-Accelerated Algorithms for Graph Vector Search: Taxonomy, Empirical Study, and Research Directions

Probability-Invariant Random Walk Learning on Gyral Folding-Based Cortical Similarity Networks for Alzheimer's and Lewy Body Dementia Diagnosis

The Correspondence Between Bounded Graph Neural Networks and Fragments of First-Order Logic

GraFSTNet: Graph-based Frequency SpatioTemporal Network for Cellular Traffic Prediction

SAGE: Structure Aware Graph Expansion for Retrieval of Heterogeneous Data

Training-free Graph-based Imputation of Missing Modalities in Multimodal Recommendation