Question Answering On Multirc
評価指標
EM
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
比較表
| モデル名 | EM |
|---|---|
| hungry-hungry-hippos-towards-language | 48.9 |
| finetuned-language-models-are-zero-shot | - |
| deberta-decoding-enhanced-bert-with | 63.7 |
| bloomberggpt-a-large-language-model-for | - |
| exploring-the-limits-of-transfer-learning | - |
| bloomberggpt-a-large-language-model-for | - |
| exploring-the-limits-of-transfer-learning | 63.3 |
| bloomberggpt-a-large-language-model-for | - |
| kelm-knowledge-enhanced-pre-trained-language | 27.2 |
| hungry-hungry-hippos-towards-language | 59.5 |
| palm-2-technical-report-1 | - |
| designing-effective-sparse-expert-models | - |
| bert-pre-training-of-deep-bidirectional | 24.1 |
| language-models-are-few-shot-learners | - |
| hungry-hungry-hippos-towards-language | 59.7 |
| palm-scaling-language-modeling-with-pathways-1 | 69.2 |
| ask-me-anything-a-simple-strategy-for | - |
| hungry-hungry-hippos-towards-language | 51.4 |
| toward-efficient-language-model-pretraining | 63 |
| ask-me-anything-a-simple-strategy-for | - |
| ask-me-anything-a-simple-strategy-for | - |
| palm-2-technical-report-1 | - |
| finetuned-language-models-are-zero-shot | - |
| n-grammer-augmenting-transformers-with-latent-1 | 11.3 |
| designing-effective-sparse-expert-models | - |
| alexatm-20b-few-shot-learning-using-a-large | - |
| bloomberggpt-a-large-language-model-for | - |
| toward-efficient-language-model-pretraining | 62.4 |
| palm-2-technical-report-1 | - |
| finetuned-language-models-are-zero-shot | - |