| GPT-4 Technical Report | ✓ Link | 72.9 | 269 | 43.4 | 29.1 | 32.8 | 7 | | GPT-4 (5-shot) | 2023-03-15 |
| GPT-4 Technical Report | ✓ Link | 73.4 | 262 | 43.7 | 29.7 | 33.5 | 4 | | GPT-4 (1-shot) | 2023-03-15 |
| GPT-4 Technical Report | ✓ Link | 73.6 | 249 | 42.8 | 28.5 | 32.3 | 3 | | GPT-4 (100-shot) | 2023-03-15 |
| GPT-4 Technical Report | ✓ Link | 73.7 | 272 | 43.9 | 29.9 | 33.6 | 5 | | GPT-4 (3-shot) | 2023-03-15 |
| GPT-4 Technical Report | ✓ Link | 75.8 | 239 | 41.5 | 27.2 | 30.7 | 6 | | GPT-4 (0-shot) | 2023-03-15 |
| GPT-4 Technical Report | ✓ Link | 80.6 | 149 | 37.3 | 22.0 | 25.4 | 2 | | GPT-3.5-turbo (5-shot) | 2023-03-15 |
| GPT-4 Technical Report | ✓ Link | 80.9 | 140 | 36.8 | 21.3 | 24.7 | 0 | | GPT-3.5-turbo (3-shot) | 2023-03-15 |
| GPT-4 Technical Report | ✓ Link | 81.2 | 137 | 36.1 | 20.4 | 24.0 | 2 | | GPT-3.5-turbo (10-shot) | 2023-03-15 |
| GPT-4 Technical Report | ✓ Link | 82.3 | 123 | 34.4 | 18.2 | 21.2 | 0 | | GPT-3.5-turbo (1-shot) | 2023-03-15 |
| GPT-4 Technical Report | ✓ Link | 82.5 | 114 | 34.0 | 18.4 | 21.6 | 0 | | GPT-3.5-turbo (0-shot) | 2023-03-15 |
| Text Embeddings by Weakly-Supervised Contrastive Pre-training | ✓ Link | 83.8 ± .6 | 89 ± 6 | 33.1 ± .3 | 16.3 ± .4 | 19.5 ± .4 | 1 ± 0 | | E5 (BASE) | 2022-12-07 |
| Learning Word Vectors for 157 Languages | ✓ Link | 84.2 ± .5 | 80 ± 4 | 32.1 ± .3 | 15.2 ± .3 | 18.4 ± .4 | 0 ± 0 | | FastText (Crawl) | 2018-02-19 |
| Text Embeddings by Weakly-Supervised Contrastive Pre-training | ✓ Link | 84.4 ± .7 | 76 ± 5 | 32.3 ± .4 | 15.4 ± .5 | 18.5 ± .6 | 0 ± 0 | | E5 (LARGE) | 2022-12-07 |
| GloVe: Global Vectors for Word Representation | ✓ Link | 84.9 ± .4 | 68 ± 4 | 31.5 ± .3 | 14.4 ± .3 | 17.6 ± .4 | 0 ± 0 | | GloVe | 2014-10-01 |
| Learning Word Vectors for 157 Languages | ✓ Link | 85.5 ± .5 | 62 ± 3 | 30.4 ± .2 | 13.0 ± .2 | 15.8 ± .3 | 0 ± 0 | | FastText (News) | 2018-02-19 |
| MPNet: Masked and Permuted Pre-training for Language Understanding | ✓ Link | 86.3 ± .4 | 50 ± 4 | 29.4 ± .3 | 11.7 ± .4 | 14.3 ± .5 | 0 ± 0 | | all-mpnet (BASE) | 2020-04-20 |
| Pre-Training of Deep Bidirectional Protein Sequence Representations with Structural Information | ✓ Link | 88.3 ± .5 | 33 ± 2 | 26.5 ± .2 | 8.2 ± .3 | 10.3 ± .3 | 0 ± 0 | | BERT (LARGE) | 2019-11-25 |
| Pre-Training of Deep Bidirectional Protein Sequence Representations with Structural Information | ✓ Link | 89.5 ± .4 | 22 ± 2 | 25.1 ± .2 | 6.4 ± .3 | 8.1 ± .4 | 0 ± 0 | | BERT (BASE) | 2019-11-25 |
| Large Language Models are Fixated by Red Herrings: Exploring Creative Problem Solving and Einstellung Effect using the Only Connect Wall Dataset | ✓ Link | | 1405 | | | | 285 | | Human Performance | 2023-06-19 |
| Deep contextualized word representations | ✓ Link | | 55 ± 4 | 29.5 ± .3 | 11.8 ± .4 | 14.5 ± .4 | 0 ± 0 | 86.3 ± .6 | ELMo (LARGE) | 2018-02-15 |
| DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter | ✓ Link | | 49 ± 4 | 29.1 ± .2 | 11.3 ± .3 | 14.0 ± .3 | 0 ± 0 | 86.7 ± .6 | DistilBERT (BASE) | 2019-10-02 |
| RoBERTa: A Robustly Optimized BERT Pretraining Approach | ✓ Link | | 29 ± 3 | 26.7 ± .2 | 8.4 ± .3 | 9.4 ± .4 | 0 ± 0 | 88.4 ± .4 | RoBERTa (LARGE) | 2019-07-26 |