Table 3.
The attention-based models in molecular property prediction research
| Tool | Attention-based model | Jointly used model or strategy | Reference |
|---|---|---|---|
| SAMPN | self-attention | MPNN | [136] |
| / | Attention | MPNN | [137] |
| MV-GNN | Attention | GCN/Multi-view Learning | [151] |
| SME | Attention | GCN | [56] |
| EAGCN | Edge Attention | GCN | [152] |
| CasANGCL | Cascaded Attention | Pre-training/Graph Contrastive Learning | [138] |
| HiGNN | Feature-Wise Attention | GNN | [49] |
| MG-BERT | BERT | GNN | [81] |
| K-BERT | BERT | Pre-training/Contrastive Learning/Fine-tuning | [80] |
| MolRoPE-BERT | BERT | Pre-training/Fine-tuning | [78] |
| FP-BERT | BERT | CNN | [143] |
| SMG-BERT | BERT | [144] | |
| ChemBERTa | BERT | Pre-training | [82] |
| ChemBERTa-2 | BERT | Pre-training | [83] |
| SMILES-BERT | BERT | Pre-training | [145] |
| FraGAT | GAT | [139] | |
| ATMOL | GAT | Graph Contrastive Learning | [57] |
| FP-GNN | GAT | [140] | |
| MoGAT | GAT | [141] | |
| PredPS | GAT | [142] | |
| ExGCN | GAT | GCN | [153] |
| ABT-MPNN | Transformer | MPNN | [147] |
| TranGRU | Transformer | BiGRU | [148] |
| DHTNN | Transformer | [149] | |
| MolHGT | Heterogeneous Graph Transformer | [146] | |
| PharmHGT | [59] | ||
| GROVER | GNN Transformer | Pre-training | [150] |