AI Paper Atlas

Research map / Graph learning / GNN architectures

Task-Agnostic Graph Explanations

Yaochen Xie, Sumeet Katariya, Xianfeng Tang, Edward W Huang, Nikhil Rao, Karthik Subbian, Shuiwang Ji

NeurIPS 2022 · Graph learning › GNN architectures · approach: GNN explainability · 4 citations (OpenAlex)

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
task-agnostic TAGE vs task-specific PGExplainer trained on same taskTAGE · MoleculeNet BACEfidelity: 0.252 -> 0.378Improved
TAGE vs PGExplainer on same taskTAGE · MoleculeNet HIVfidelity: 0.473 -> 0.595Improved
TAGE vs PGExplainer on same taskTAGE · MoleculeNet BBBPfidelity: 0.182 -> 0.193No clear effect
TAGE vs PGExplainer on same taskTAGE · PPI Task 2fidelity: 0.345 -> 0.499Improved
TAGE vs PGExplainer on same taskTAGE · E-commerce Sellersfidelity: 0.3246 -> 0.6515Improved
evaluate PGExplainer on a task different from its training taskPGExplainer · MoleculeNet BACEfidelity: 0.252 -> 0.007Hurt
evaluate PGExplainer on a task different from its training taskPGExplainer · MoleculeNet HIVfidelity: 0.473 -> -0.001Hurt
train TAGE on individual MoleculeNet dataset instead of ZINC (ablation)TAGE · MoleculeNet BACEfidelity: 0.378 -> 0.402Mixed
train TAGE on individual MoleculeNet dataset instead of ZINC (ablation)TAGE · MoleculeNet HIVfidelity: 0.595 -> 0.541Mixed
TAGE trained on individual datasets vs PGExplainer (ablation)TAGE · MoleculeNet SIDERfidelity: 0.444 -> 0.516Improved
TAGE vs GEM causality-based explainerTAGE · BACE sparsity 0.85fidelity: 0.3607 -> 0.4992Improved
TAGE vs PGExplainer at lower sparsityTAGE · BACE sparsity 0.75fidelity: 0.5161 -> 0.5309No clear effect

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