AI Paper Atlas

Research map / Graph learning / GNN architectures

Rethinking Link Prediction for Directed Graphs

Mingguo He, Yuhe Guo, Yu Zheng, Zhewei Wei, Stephan Günnemann, Xiaokui Xiao

TPAMI 2026 · 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
compare simple MLP against DUPLEX under existing setupMLP · Cora (DUPLEX setup, test edges in training)DP AUC: DUPLEX 96.34 vs MLP 99.25 DP AUC, but DUPLEX better on 3C ACC 92.41 vs 79.97Mixed
remove test edges from training graph (fix label leakage)DUPLEX · Cora (DUPLEX setup)EP ACC: 93.49 -> 87.43Hurt
remove test edges from training graph (fix label leakage)DUPLEX · CiteSeer (DUPLEX setup)EP ACC: 92.11 -> 83.59Hurt
add polynomial graph filter (SDGAE) in place of DiGAE convolution (ablation)DiGAE · DirLinkBench Cora-MLHits@100: 82.06 -> 90.37Improved
add polynomial graph filter (SDGAE) in place of DiGAE convolution (ablation)DiGAE · DirLinkBench CiteSeerHits@100: 83.64 -> 93.69Improved
add polynomial graph filter (SDGAE) in place of DiGAE convolution (ablation)DiGAE · DirLinkBench PhotoHits@100: 55.05 -> 68.84Improved
use symmetric normalization on DiGAE (ablation)DiGAE · DirLinkBench Cora-MLHits@100: 82.06 -> 70.89Hurt
use symmetric normalization on DiGAE (ablation)DiGAE · DirLinkBench CiteSeerHits@100: 83.64 -> 71.60Hurt
increase polynomial order K (ablation)SDGAE · Cora-ML and CiteSeerHits@100: consistently improves with increasing K, optimal at K = 5Improved
increase number of convolution layers L (ablation)DiGAE · Cora-ML and CiteSeerHits@100: performance declines as layers increaseHurt
add residual connections (ablation)DiGAE · Cora-ML and CiteSeerHits@100: simply adding residual connections fails to improve DiGAE's performanceNo clear effect
use BCE loss instead of CE loss (ablation)MagNet · DirLinkBenchHits@100: BCE consistently outperforms CEImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

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