← Back to all news
arXiv Machine Learning By Asela Hevapathige, Ahad N. Zehmakan, Asiri Wijesinghe, Saman Halgamuge

Not Just Oversmoothing: Detecting the Echo Chamber Effect in Graph Neural Networks

Read the original on arXiv Machine Learning →

The paper introduces the Echo Chamber Effect, a failure mode in Graph Neural Networks where intra-community representations collapse while inter-community separation remains, differing from traditional oversmoothing. It proposes the Echo Chamber Index (ECI) to detect this effect by stratifying pairwise distances by community membership. Building on this analysis, the authors present Community-Aware Split Propagation (CASP), a lightweight plugin that decouples intra- and inter-community aggregation and learns their balance from label structure, improving performance across various GNN backbones in both homophilic and heterophilic settings.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

arXiv Machine Learning
Sep 23

Diffusion-Induced Spatial Attention Overlapping Community Detection

The paper introduces DISCO, a deep‑learning framework for detecting overlapping communities in networks. DISCO integrates a diffusion‑based structural prior, sparse multi‑head attention, and a Bernoulli‑Poisson edge‑reconstruction objective to infer community affiliations from node attributes and structural profiles. Experiments show competitive performance against existing graph convolutional and attention methods, and a cybersecurity proof‑of‑concept demonstrates how community changes can signal anomalies in dynamic communication networks.

By Kosti Koistinen, Vesa Kuikka, Joni Herttuainen, Matthew Hendren, Brian Holt, Kimmo K. Kaski
diffusionbenchmarks
More like this →
arXiv Machine Learning
Aug 19

Community Concealment from Graph Neural Networks

The paper introduces FCom‑DICE, a feature‑aware perturbation method that rewires influential edges and adjusts node features to hide a target community from graph neural network (GNN) inference. It shows that concealment effectiveness depends on boundary connectivity and feature similarity, and that FCom‑DICE outperforms structure‑only DICE on synthetic and real networks such as Facebook, Wikipedia, and Bitcoin Transactions while preserving key structural and feature properties.

By Dalyapraz Manatova, Pablo Moriano, L. Jean Camp
benchmarkssafety
More like this →