Fig 1.
Overview of the proposed MSG-Pre framework.
The framework consists of three main components: (1) scale-adaptive attention, which dynamically adjusts the importance of geometric relationships based on their spatial context; (2) hierarchical contrastive learning, which encourages the model to learn meaningful representations by contrasting similar and dissimilar molecules; and (3) geometric regularization, which enforces physical constraints on the learned representations.
Table 1.
Dataset statistics with training/validation/test splits.
Table 2.
Performance comparison on representative benchmarks with 95% confidence intervals.
Table 3.
Performance on complex property prediction tasks.
Table 4.
Cross-domain generalization results (ROC-AUC).
Table 5.
Ablation study on key components (QM9 MAE ↓).
Fig 2.
Multi-scale attention visualization on Apixaban (PDB ID: 4yhy).
Atomic (s1), group (s2), and conformer (s3) attention weights are shown in red, blue, and green, respectively.
Table 6.
Mutual information analysis (Normalized Scores ↑).