AI Paper Atlas

Research map / Graph learning / Knowledge and temporal graphs

Dual-channel Dynamic Graph Neural Networks with Adaptive Adjacency Learning and Multi-scale Representation Fusion

Youqing Wang, Jiahao Long, Tianxiang Zhao, Man Cao, Mengyuan Xin, Jiapu Wang, Junbin Gao, Jipeng Guo

ICML 2026 · Graph learning › Knowledge and temporal graphs · approach: Temporal graph neural architectures

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
add dynamic semantic adjacency channel (ablation)DCD-GNN · Wisconsinnode classification accuracy (%): 77.00 -> 98.13Improved
add dynamic semantic adjacency channel (ablation)DCD-GNN · Texasnode classification accuracy (%): 78.85 -> 94.10Improved
add dynamic semantic adjacency channel (ablation)DCD-GNN · Chameleonnode classification accuracy (%): 69.04 -> 78.14Improved
add dynamic semantic adjacency channel (ablation)DCD-GNN · Coranode classification accuracy (%): 88.92 -> 90.34Improved
add static graph channel (ablation)DCD-GNN · Chameleonnode classification accuracy (%): 75.65 -> 78.14Improved
add static graph channel (ablation)DCD-GNN · Texasnode classification accuracy (%): 92.95 -> 94.10Improved
add static graph channel (ablation)DCD-GNN · Actornode classification accuracy (%): 40.91 -> 41.04No clear effect
add multi-scale representation fusion (ablation)DCD-GNN · Texasnode classification accuracy (%): 80.16 -> 94.10Improved
add multi-scale representation fusion (ablation)DCD-GNN · Wisconsinnode classification accuracy (%): 92.75 -> 98.13Improved
add multi-scale representation fusion (ablation)DCD-GNN · Citeseernode classification accuracy (%): 80.36 -> 82.43Improved
add multi-scale representation fusion (ablation)DCD-GNN · Actornode classification accuracy (%): 41.00 -> 41.04No clear effect
increase network depth from 2 to 6 layersDCD-GNN · Wisconsinnode classification accuracy (%): 98.13 -> 92.50Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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