AI Paper Atlas

Research map / Graph learning / Graph self-supervised learning

Hypergraph Clustering Network with Partial Attribute Imputation

Qianqian Wang, Bowen Zhao, Zhengming Ding, Wei Feng, Quanxue Gao

ICCV 2025 · Graph learning › Graph self-supervised learning · approach: Heterophily and imbalance on graphs · 5 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
Add higher-order neighborhood propagation (Dirichlet energy imputation) (ablation)HCN-PAI · CORA (missing rate 0.6)ACC (%): 49.55 -> 68.03Improved
Add higher-order neighborhood propagation (ablation)HCN-PAI · ModelNet40 (missing rate 0.6)ACC (%): 55.93 -> 85.47Improved
Add dual-level contrastive mechanism (ablation)HCN-PAI · CORA (missing rate 0.6)ACC (%): 62.98 -> 68.03Improved
Add hypergraph smoothing preprocessing (ablation)HCN-PAI · CITESEER (missing rate 0.6)ACC (%): 58.36 -> 62.16Improved
Add hypergraph smoothing preprocessing (ablation)HCN-PAI · NTU2012 (missing rate 0.6)ACC (%): 66.32 -> 67.48Improved
Use higher-order neighborhood propagation instead of zero, random or mean imputation (ablation)HCN-PAI · Four datasets (missing rate 0.9)clustering metricsImproved
Replace hypergraph convolution with smoothing preprocessing and unshared MLPsHCN-PAI · CORA training timeseconds: 470.06 (MMACL) -> 2.49Improved
Vary trade-off parameter beta over 0.01 to 100 (ablation)HCN-PAI · CORA, CITESEERclustering metricsNo clear effect
Vary smoothing parameter r on ModelNet40 (ablation)HCN-PAI · ModelNet40ACCHurt
Increase missing rate from 0.3 to 0.9 on CORAHCN-PAI · CORAACC (%): 72.51 -> 66.05Hurt

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