AI Paper Atlas

Research map / Graph learning / Graph self-supervised learning

HINPool: A Unified Heterogeneous Graph Pooling Framework for Accurate Molecular and Protein Property Prediction

Ming-Yi Hong, You-Chen Teng, Shao-En Lin, Chih-Yu Wang, Che Lin

AAAI 2026 · Graph learning › Graph self-supervised learning · approach: Molecular graph pretraining · 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
add cross-layer fusion (ablation)HINPool · MUTAGAUROC: 60.8 -> 100.0Improved
add cross-layer fusion (ablation)HINPool · ENZYMESAUROC: 71.5 -> 92.8Improved
add type-aware selector (TAS) (ablation)HINPool · PROTEINSAUROC: 86.9 -> 89.8Improved
use type-aware Top-K instead of mixed Top-K (ablation)HINPool · ENZYMESAUROC: 92.7 -> 92.8No clear effect
add type-aware readout fusion (ablation)HINPool · ENZYMESAUROC: 73.2 -> 92.8Improved
model data as HIN with HINPool vs best baselineHINPool · NCI109AUROC: 86.5 -> 86.2Hurt
model data as HIN with HINPool vs best baselineHINPool · ENZYMESAUROC: 86.2 -> 92.8Improved
reduce pooling ratio from 0.9 to 0.6-0.8 (ablation)HINPool · PROTEINSAUROC: 89.8 -> 88.3 (0.8), 88.6 (0.7), 88.5 (0.6)Hurt
use 3 layers instead of 2 or 4 (ablation)HINPool · PROTEINSAUROC: 87.9 / 87.7 -> 89.8Improved
use declining pooling ratios 0.9/0.8/0.7 (ablation)HINPool · PROTEINSAUROC: 89.8 -> 88.4Hurt

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