AI Paper Atlas

Research map / Graph learning / GNN architectures

V-InFoR: A Robust Graph Neural Networks Explainer for Structurally Corrupted Graphs

Senzhang Wang, Jun Yin, Chaozhuo Li, Xing Xie, Jianxin Wang

NeurIPS 2023 · Graph learning › GNN architectures · approach: Graph structure learning · 0 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
Use V-InfoR instead of best baseline under 20% random noiseGNN explainers · MutagFNS: 0.2207 -> 0.2852Improved
Use V-InfoR instead of best baseline under 20% random noiseGNN explainers · BA-3MotifsFNS: 0.4222 -> 0.4610Improved
Use V-InfoR instead of best baseline under 20% random noiseGNN explainers · Ogbg-molhivFNS: 0.2198 -> 0.2542Improved
Use V-InfoR instead of best baseline under 20% random noiseGNN explainers · Ogbg-ppaFNS: 0.5200 -> 0.5680Improved
Use V-InfoR instead of best baseline under GRABNEL attack, 5% budgetGNN explainers · BA-3MotifsFNS: 0.5582 -> 0.6157Improved
Use V-InfoR instead of best baseline under GRABNEL attack, 10% budgetGNN explainers · MutagFNS: 0.4904 -> 0.5427Improved
Raise GRABNEL attack budget from 5% to 10%V-InfoR · MutagPN: 0.4588 -> 0.4277Mixed
Add noise ratio 0 to 0.30 sweep (no corruption)V-InfoR · BA-3Motifs and MutagFNSImproved
Remove the robust graph representation extractor (No-VAE) (ablation)V-InfoR · BA-3Motifs, Mutag, etc.PS/PN/FNSImproved
Replace GIB objective with sparsity-driven MI objective (No-GIB) (ablation)V-InfoR · all datasetsFNSImproved
Remove both VAE extractor and GIB (No-VAE-GIB) (ablation)V-InfoR · all datasetsFNSImproved
Vary tau in [0.1, 0.5]V-InfoR · BA-3Motifs and MutagFNSImproved

Later papers that compared with this method

1 later accepted paper reports V-InFoR as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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