AI Paper Atlas

Research map / Graph learning / Graph algorithms and clustering

Rethinking the Gold Standard: Why Discrete Curvature Fails to Fully Capture Over-squashing in GNNs?

Jialong Chen, Bowen Deng, Zibin Zheng, Chuan Chen

ICLR 2026 · Graph learning › Graph algorithms and clustering · approach: Graph comparison and random features

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use Augmented Forman-3 instead of Ollivier Ricci curvatureGCN · 21 datasets averageMOSR10: .271 -> .067Improved
use Balanced Forman instead of Ollivier Ricci curvatureGCN · 21 datasets averageMOSR10: .271 -> .078Improved
add degree weighting f=1/(1+x) to AF3 (WAF3)GCN · 21 datasets averageMOSR10: .067 -> .036Improved
add degree weighting f=1/(1+x) to AF3 (WAF3)GAT · 21 datasets averageMOSR10: .148 -> .111Improved
add degree weighting f=1/(1+x) to AF3 (WAF3)GraphSAGE · 21 datasets averageMOSR25: .161 -> .124Improved
replace GCN with GAT or GraphSAGE as GNNOllivier Ricci · 21 datasets averageMOSR10: .271 -> .336 -> .386Hurt
raise q from 10 to 25Ollivier Ricci GCN · 21 datasets averageMOSR: .271 -> .307Hurt
use MinHash approximation with H=100 instead of exact WAF3WAF3 · Erdos-Renyi 5M edgescomputation time: 3142s -> 23.5s, 133.7x speed upImproved
raise MinHash hashes from H=100 to H=1000 (ablation)WAF3 approximation · 21 datasetsKendall tau-b: 95% -> over 98%Improved
raise MinHash hashes from H=1000 to H=10000 (ablation)WAF3 approximation · 21 datasetsKendall tau-b: less than 1% improvementNo clear effect
train GNN for more epochs before measuring MOSR (ablation)GNNs · Citeseer, Cora, PubmedMOSR: slowly increases then stabilisesHurt
use f=(1+d)^-1 instead of (1+d)^-2 or (1+d)^-1/2 (ablation)WAF3 · MOSR across datasetsMOSR: slightly smaller on averageImproved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy