Cross Modal Retrieval On Chebi 20
评估指标
Hits@1
Hits@10
Mean Rank
Test MRR
评测结果
各个模型在此基准测试上的表现结果
模型名称 | Hits@1 | Hits@10 | Mean Rank | Test MRR | Paper Title | Repository |
|---|---|---|---|---|---|---|
| GCN2 | 22.3 | 68.9 | 41.90 | 37.1 | Text2Mol: Cross-Modal Molecule Retrieval with Natural Language Queries | |
| CLASS (ORMA) | 67.4 | 93.4 | 17.82 | 77.4 | CLASS: Enhancing Cross-Modal Text-Molecule Retrieval Performance and Training Efficiency | - |
| AMAN | 49.4 | 92.1 | 16.01 | 64.7 | Adversarial Modality Alignment Network for Cross-Modal Molecule Retrieval | |
| Song et al. | 56.5 | 94.1 | 12.66 | 70.2 | Towards Cross-Modal Text-Molecule Retrieval with Better Modality Alignment | |
| DSOKR | 51.0 | 88.2 | 76.43 | 64.2 | Deep Sketched Output Kernel Regression for Structured Prediction | |
| ORMA | 66.5 | 93.9 | 18.53 | 77.2 | Exploring Optimal Transport-Based Multi-Grained Alignments for Text-Molecule Retrieval | - |
| All-Ensemble | 34.4 | 81.1 | 20.21 | 49.9 | Text2Mol: Cross-Modal Molecule Retrieval with Natural Language Queries | |
| MLP1 | 22.4 | 68.6 | 30.38 | 37.2 | Text2Mol: Cross-Modal Molecule Retrieval with Natural Language Queries | |
| CLASS (AMAN) | 51.1 | 92.6 | 16.80 | 66.0 | CLASS: Enhancing Cross-Modal Text-Molecule Retrieval Performance and Training Efficiency | - |
0 of 9 row(s) selected.