Text+Chem T5-Base

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· 2023-01-29
58
best: 73.2 (Mol-LLM (Mistral-Instruct-v0.2))
Unifying Molecular and Textual Representations via Multi-task Language Modelling arXiv:2301.12586
Molecule CaptioningonChEBI-20
BLEU-4· 2023-01-29
49
best: 60.8 (MolReFlect)
Unifying Molecular and Textual Representations via Multi-task Language Modelling arXiv:2301.12586
Molecule CaptioningonChEBI-20
METEOR· 2023-01-29
60.4
best: 68.1 (BioT5+)
Unifying Molecular and Textual Representations via Multi-task Language Modelling arXiv:2301.12586
Molecule CaptioningonChEBI-20
ROUGE-1· 2023-01-29
64.7
best: 71 (BioT5+)
Unifying Molecular and Textual Representations via Multi-task Language Modelling arXiv:2301.12586
Molecule CaptioningonChEBI-20
ROUGE-2· 2023-01-29
49.8
best: 61.8 (XMolCap)
Unifying Molecular and Textual Representations via Multi-task Language Modelling arXiv:2301.12586
Molecule CaptioningonChEBI-20
ROUGE-L· 2023-01-29
58.6
best: 65.3 (PEIT-GEN)
Unifying Molecular and Textual Representations via Multi-task Language Modelling arXiv:2301.12586