EstGraph benchmark evaluates LLMs on estimating properties of very large graphs from random-walk samples that fit in context limits.
A comprehensive survey on graph neural networks
8 Pith papers cite this work. Polarity classification is still indexing.
8
Pith papers citing it
representative citing papers
Small instruction-tuned language models cannot reliably estimate graph-theoretic properties from textual encodings, though adjacency-list formats and multi-branch reasoning reduce errors relative to edge lists and single-path inference.
A graph neural network recommends combinatorial choices for tether-net morphology, masses, thrusters, and aiming points, reducing the MCNLP to an NLP solved by PSO and yielding faster convergence than direct optimization.
IGCNet learns power control policies for interference channels via graph convolutions, is proven to be a universal approximator for permutation-invariant continuous functions, and outperforms WMMSE in speed while remaining robust to imperfect CSI.
ML-GCN embeds nodes via GCN, generates a label matrix in the same space, and trains with relaxed skip-gram on node-label concatenations to model correlations, reporting outperformance on graph datasets.
Presents YoutubeGraph-Dyn, a multi-modal dynamic graph dataset from YouTube interactions with intra-day snapshots, and benchmarks clustering for community migration plus time series and RNN methods for forecasting non-timestamped attributes.
A spectral-based GCN for directed graphs uses redefined Laplacians to enable direct application to directed data and outperforms prior methods on semi-supervised node classification tasks.
Techniques enable training the sparse GNN from Allamanis et al. [2018] on dense TPU hardware in 13 minutes versus a full day originally.
citing papers explorer
-
Evaluating LLMs on Large-Scale Graph Property Estimation via Random Walks
EstGraph benchmark evaluates LLMs on estimating properties of very large graphs from random-walk samples that fit in context limits.
-
Graph Property Inference in Small Language Models: Effects of Representation and Reasoning Strategy
Small instruction-tuned language models cannot reliably estimate graph-theoretic properties from textual encodings, though adjacency-list formats and multi-branch reasoning reduce errors relative to edge lists and single-path inference.
-
Designing Active Tether-Net Systems for Space Debris Capture with Graph-Learning-Aided Mixed-Combinatorial Optimization
A graph neural network recommends combinatorial choices for tether-net morphology, masses, thrusters, and aiming points, reducing the MCNLP to an NLP solved by PSO and yielding faster convergence than direct optimization.
-
A Graph Neural Network Approach for Scalable Wireless Power Control
IGCNet learns power control policies for interference channels via graph convolutions, is proven to be a universal approximator for permutation-invariant continuous functions, and outperforms WMMSE in speed while remaining robust to imperfect CSI.
-
Semi-Supervised Graph Embedding for Multi-Label Graph Node Classification
ML-GCN embeds nodes via GCN, generates a label matrix in the same space, and trains with relaxed skip-gram on node-label concatenations to model correlations, reporting outperformance on graph datasets.
-
Tracking Temporal Evolution of Graphs using Non-Timestamped Data
Presents YoutubeGraph-Dyn, a multi-modal dynamic graph dataset from YouTube interactions with intra-day snapshots, and benchmarks clustering for community migration plus time series and RNN methods for forecasting non-timestamped attributes.
-
Spectral-based Graph Convolutional Network for Directed Graphs
A spectral-based GCN for directed graphs uses redefined Laplacians to enable direct application to directed data and outperforms prior methods on semi-supervised node classification tasks.
-
Fast Training of Sparse Graph Neural Networks on Dense Hardware
Techniques enable training the sparse GNN from Allamanis et al. [2018] on dense TPU hardware in 13 minutes versus a full day originally.