arXiv Machine Learning July 1, 2026 By Tobias Schumacher, Simon Reichelt, Markus Strohmaier

The Impact of Dimensionality on the Stability of Node Embeddings

Read the original on arXiv Machine Learning →

arXiv:2604. 08492v2 Announce Type: replace Abstract: Previous work has shown that node embedding methods can produce different representations and downstream predictions across repeated training runs, even when trained on the same data with identical hyperparameters.

Summary generated by The Flow from the publisher's feed. The full article lives at arXiv Machine Learning.

arXiv Machine Learning

Jun 3

Link Prediction or Perdition: the Seeds of Instability in Knowledge Graph Embeddings

arXiv:2606. 03365v1 Announce Type: new Abstract: Embedding models (KGEMs) constitute the main link prediction approach to complete knowledge graphs.

By Guillaume M\'erou\'e, Fabien Gandon, Pierre Monnin

rag benchmarks

arXiv AI

Jun 2

On the Theoretical Limitations of Embedding-based Link Prediction

arXiv:2506. 22271v3 Announce Type: replace Abstract: Neural networks often map low-dimensional embeddings to high-dimensional output spaces.

By Samy Badreddine, Emile van Krieken, Luciano Serafini

rag

arXiv AI

Aug 11

LEED: Local Embedding Evolution Distance for over-smoothing estimation and virtual node selection in GNN

arXiv:2608. 09596v1 Announce Type: cross Abstract: Graph Neural Networks (GNNs) suffer from two fundamental limitations: over-smoothing, where node representations become indistinguishable with depth, and over-squashing, where long-range information is compressed through limited message-passing channels.

By Killian Cressant, Pedro B. Velloso

rag

arXiv Machine Learning

Jul 21

Node4All: Learning Node Representation Beyond Datasets

arXiv:2607. 17272v1 Announce Type: new Abstract: Node representation learning has advanced rapidly, yet most existing methods rely on per-dataset training and hyperparameter tuning.

By Dooho Lee, Jaemin Yoo

llms benchmarks

arXiv Machine Learning

Jun 2

Towards Stable, Globally Expressive Graph Representations with Laplacian Eigenvectors

arXiv:2410. 09737v2 Announce Type: replace Abstract: A popular way to improve the expressive power of graph neural networks (GNNs) is to use Laplacian eigenvectors as additional node features, since they can serve both as structural identifiers and global coordinates of nodes.

By Junru Zhou, Cai Zhou, Xiyuan Wang, Pan Li, Muhan Zhang

benchmarks

arXiv AI

Jun 11

Unstable Features, Reproducible Subspaces: Understanding Seed Dependence in Sparse Autoencoders

arXiv:2606. 12138v1 Announce Type: cross Abstract: Sparse autoencoders (SAEs) are widely used to interpret neural network representations, but their utility depends on whether the learned features are reproducible across training runs.

By Gleb Gerasimov, Timofei Rusalev, Nikita Balagansky, Daniil Laptev, Vadim Kurochkin, Daniil Gavrilov

Related stories

Link Prediction or Perdition: the Seeds of Instability in Knowledge Graph Embeddings

On the Theoretical Limitations of Embedding-based Link Prediction

LEED: Local Embedding Evolution Distance for over-smoothing estimation and virtual node selection in GNN

Node4All: Learning Node Representation Beyond Datasets

Towards Stable, Globally Expressive Graph Representations with Laplacian Eigenvectors

Unstable Features, Reproducible Subspaces: Understanding Seed Dependence in Sparse Autoencoders