AI Paper Atlas

Research map / Graph learning / GNN architectures

BiKT: Unleashing the Potential of GNNs via Bi-Directional Knowledge Transfer

Shuai Zheng, Zhizhe Liu, Zhenfeng Zhu, Xingxing Zhang, Jianxin Li, Yao Zhao

TPAMI 2025 · Graph learning › GNN architectures · approach: Simple and scalable GNNs

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 BiKT bi-directional knowledge transferGCN · Cora (transductive)accuracy: 73.44 -> 76.91Improved
add BiKTGCN · Citeseer (transductive)accuracy: 65.97 -> 68.63Improved
add BiKTGCN · Pubmed (transductive)accuracy: 81.62 -> 82.28Improved
add BiKTMixHop · Citeseer (transductive)accuracy: 62.64 -> 68.28Improved
add BiKTGraphSAGE · OGB-Arxiv (transductive)accuracy: 74.55 -> 75.47Improved
add BiKTGraphSAGE · OGB-Products (inductive)accuracy: 76.98 -> 77.47Improved
add BiKTGCN · MUTAGaccuracy: 84.16 -> 87.26Improved
add BiKTGIN · OGB-molhivaccuracy: 76.09 -> 76.47No clear effect
train MLP with BiKT instead of plain MLP_GCNMLP_GCN · Cora (transductive)accuracy: 49.84 -> 74.99Improved
use BiKT-MLP_GCN instead of BiKT-GCNBiKT-GCN · Citeseer (transductive)accuracy: 68.63 -> 69.61Improved
add BiKT on disassortative nodesGCN · Citeseer disassortative nodesaccuracy: 52.33 -> 65.14Improved
combine knowledge transfer and knowledge distillation terms (ablation)BiKT · Cora and CiteseeraccuracyImproved

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