Common Sense Reasoning On Record
المقاييس
EM
F1
النتائج
نتائج أداء النماذج المختلفة على هذا المعيار القياسي
جدول المقارنة
| اسم النموذج | EM | F1 |
|---|---|---|
| النموذج 1 | 81.460 | 82.664 |
| luke-graph-a-transformer-based-approach-with | 91.2 | 91.5 |
| finetuned-language-models-are-zero-shot | 72.5 | - |
| record-bridging-the-gap-between-human-and | 45.4 | 46.7 |
| النموذج 5 | 81.780 | 82.584 |
| finetuned-language-models-are-zero-shot | 85.1 | - |
| designing-effective-sparse-expert-models | 88.9 | - |
| exploring-the-limits-of-transfer-learning | 93.4 | - |
| النموذج 9 | 59.410 | 61.515 |
| deberta-decoding-enhanced-bert-with | 94.1 | 94.5 |
| efficient-language-modeling-with-sparse-all | 79.9 | - |
| palm-scaling-language-modeling-with-pathways-1 | 94.0 | 94.6 |
| exploring-the-limits-of-transfer-learning | - | 94.1 |
| efficient-language-modeling-with-sparse-all | 60.7 | - |
| toward-efficient-language-model-pretraining | 93.9 | 94.4 |
| النموذج 16 | 83.090 | 83.737 |
| large-language-models-are-zero-shot-reasoners | - | 90.2 |
| efficient-language-modeling-with-sparse-all | 72.4 | - |
| النموذج 19 | 69.490 | 71.138 |
| language-models-are-few-shot-learners | 82.1 | - |
| bloomberggpt-a-large-language-model-for | - | 82.5 |
| efficient-language-modeling-with-sparse-all | 67.2 | - |
| النموذج 23 | 90.640 | 91.209 |
| النموذج 24 | 71.600 | 73.620 |
| palm-2-technical-report-1 | - | 93.8 |
| kelm-knowledge-enhanced-pre-trained-language | 76.2 | 76.7 |
| النموذج 27 | 59.860 | 61.885 |
| luke-deep-contextualized-entity | 90.6 | 91.2 |
| palm-2-technical-report-1 | - | 92.4 |
| toward-efficient-language-model-pretraining | 95.9 | 96.4 |
| bloomberggpt-a-large-language-model-for | - | 82.8 |
| alexatm-20b-few-shot-learning-using-a-large | - | 88.4 |
| efficient-language-modeling-with-sparse-all | 73.4 | - |
| n-grammer-augmenting-transformers-with-latent-1 | 28.9 | 29.9 |
| bloomberggpt-a-large-language-model-for | - | 67.9 |
| designing-effective-sparse-expert-models | 95.1 | - |
| kelm-knowledge-enhanced-pre-trained-language | 89.1 | 89.6 |
| النموذج 38 | 79.480 | 80.038 |
| bloomberggpt-a-large-language-model-for | - | 78 |
| pingan-smart-health-and-sjtu-at-coin-shared | 81.5 | 82.7 |
| النموذج 41 | 60.800 | 62.986 |
| palm-2-technical-report-1 | - | 92.1 |
| integrating-a-heterogeneous-graph-with-entity | 91.7 | 92.2 |
| bert-pre-training-of-deep-bidirectional | 54.040 | 56.065 |
| النموذج 45 | 72.240 | 72.778 |