PEIT-GEN

Reported on 6 benchmarks across 1 task · 1 paper · 1 SOTA

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

Medical6 results

Molecule CaptioningonChEBI-20
ROUGE-L· 2024-12-24
65.3
SOTA
Property Enhanced Instruction Tuning for Multi-task Molecule Generation with Large Language Models arXiv:2412.18084
Molecule CaptioningonChEBI-20
BLEU-2· 2024-12-24
59.8
best: 73.2 (Mol-LLM (Mistral-Instruct-v0.2))
Property Enhanced Instruction Tuning for Multi-task Molecule Generation with Large Language Models arXiv:2412.18084
Molecule CaptioningonChEBI-20
BLEU-4· 2024-12-24
53.4
best: 60.8 (MolReFlect)
Property Enhanced Instruction Tuning for Multi-task Molecule Generation with Large Language Models arXiv:2412.18084
Molecule CaptioningonChEBI-20
METEOR· 2024-12-24
67.6
best: 68.1 (BioT5+)
Property Enhanced Instruction Tuning for Multi-task Molecule Generation with Large Language Models arXiv:2412.18084
Molecule CaptioningonChEBI-20
ROUGE-1· 2024-12-24
70
best: 71 (BioT5+)
Property Enhanced Instruction Tuning for Multi-task Molecule Generation with Large Language Models arXiv:2412.18084
Molecule CaptioningonChEBI-20
ROUGE-2· 2024-12-24
58.2
best: 61.8 (XMolCap)
Property Enhanced Instruction Tuning for Multi-task Molecule Generation with Large Language Models arXiv:2412.18084