Molecule Captioning On L M 24
Metriken
BLEU-2
BLEU-4
METEOR
ROUGE-1
ROUGE-2
ROUGE-L
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
Modellname | BLEU-2 | BLEU-4 | METEOR | ROUGE-1 | ROUGE-2 | ROUGE-L | Paper Title | Repository |
|---|---|---|---|---|---|---|---|---|
| MolT5-Small | 70.9 | 51.2 | 70.1 | 74.5 | 55.8 | 54.4 | Translation between Molecules and Natural Language | |
| Nach0 | 73.81 | 53.04 | 77.45 | 80.06 | 60.17 | 57.5 | - | - |
| Mol2Lang-VLM | 77.7 | 56.3 | 74.1 | 78.6 | 59.1 | 56.5 | Mol2Lang-VLM: Vision- and Text-Guided Generative Pre-trained Language Models for Advancing Molecule Captioning through Multimodal Fusion | |
| MolT5-Base | 73.8 | 53.5 | 71.8 | 75.0 | 55.9 | 53.9 | Translation between Molecules and Natural Language | |
| MolT5-Large | 76.9 | 55.6 | 74.3 | 77.7 | 58.0 | 55.7 | Translation between Molecules and Natural Language |
0 of 5 row(s) selected.