MolCA, Galac1.3B

Reported on 6 benchmarks across 1 task · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Medical6 results

Molecule CaptioningonChEBI-20
BLEU-2· 2023-10-19
62
best: 73.2 (Mol-LLM (Mistral-Instruct-v0.2))
MolCA: Molecular Graph-Language Modeling with Cross-Modal Projector and Uni-Modal Adapter arXiv:2310.12798
Molecule CaptioningonChEBI-20
BLEU-4· 2023-10-19
53.1
best: 60.8 (MolReFlect)
MolCA: Molecular Graph-Language Modeling with Cross-Modal Projector and Uni-Modal Adapter arXiv:2310.12798
Molecule CaptioningonChEBI-20
METEOR· 2023-10-19
65.1
best: 68.1 (BioT5+)
MolCA: Molecular Graph-Language Modeling with Cross-Modal Projector and Uni-Modal Adapter arXiv:2310.12798
Molecule CaptioningonChEBI-20
ROUGE-1· 2023-10-19
68.1
best: 71 (BioT5+)
MolCA: Molecular Graph-Language Modeling with Cross-Modal Projector and Uni-Modal Adapter arXiv:2310.12798
Molecule CaptioningonChEBI-20
ROUGE-2· 2023-10-19
53.7
best: 61.8 (XMolCap)
MolCA: Molecular Graph-Language Modeling with Cross-Modal Projector and Uni-Modal Adapter arXiv:2310.12798
Molecule CaptioningonChEBI-20
ROUGE-L· 2023-10-19
61.8
best: 65.3 (PEIT-GEN)
MolCA: Molecular Graph-Language Modeling with Cross-Modal Projector and Uni-Modal Adapter arXiv:2310.12798