Metric: Accuracy (higher is better)
| # | Model↕ | Accuracy▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | BioLinkBERT (large) | 94.8 | No | LinkBERT: Pretraining Language Models with Docum... | 2022-03-29 | Code |
| 2 | GAL 120B (zero-shot) | 94.3 | No | Galactica: A Large Language Model for Science | 2022-11-16 | Code |
| 3 | BioLinkBERT (base) | 91.4 | No | LinkBERT: Pretraining Language Models with Docum... | 2022-03-29 | Code |
| 4 | BLOOM (zero-shot) | 91.4 | No | Galactica: A Large Language Model for Science | 2022-11-16 | Code |
| 5 | PubMedBERT uncased | 87.56 | No | Domain-Specific Language Model Pretraining for B... | 2020-07-31 | Code |
| 6 | GPT-4 | 85.71 | No | - | - | - |
| 7 | OPT (zero-shot) | 81.4 | No | Galactica: A Large Language Model for Science | 2022-11-16 | Code |