Common Sense Reasoning On Arc Challenge
المقاييس
Accuracy
النتائج
نتائج أداء النماذج المختلفة على هذا المعيار القياسي
جدول المقارنة
| اسم النموذج | Accuracy |
|---|---|
| model-card-and-evaluations-for-claude-models | 91 |
| large-language-models-can-self-improve | 88.3 |
| gpt-4-technical-report-1 | 96.4 |
| parameter-efficient-sparsity-crafting-from | 65.2 |
| النموذج 5 | 91.03 |
| mixture-of-subspaces-in-low-rank-adaptation | 81.5 |
| large-language-models-can-self-improve | 85.2 |
| glam-efficient-scaling-of-language-models | 50.3 |
| large-language-models-can-self-improve | 87.1 |
| designing-effective-sparse-expert-models | 86.5 |
| palm-2-technical-report-1 | 59.6 |
| galactica-a-large-language-model-for-science-1 | 32.9 |
| galactica-a-large-language-model-for-science-1 | 67.9 |
| galactica-a-large-language-model-for-science-1 | 31.1 |
| mixlora-enhancing-large-language-models-fine | 58.1 |
| massive-language-models-can-be-accurately | 25.6 |
| glam-efficient-scaling-of-language-models | 48.2 |
| mixlora-enhancing-large-language-models-fine | 69.9 |
| mixlora-enhancing-large-language-models-fine | 79.9 |
| palm-2-technical-report-1 | 95.1 |
| massive-language-models-can-be-accurately | 43.94 |
| language-models-are-few-shot-learners | 51.4 |
| large-language-models-can-self-improve | 89.8 |
| palm-2-technical-report-1 | 64.9 |
| unifying-language-learning-paradigms | 49.5 |
| llama-open-and-efficient-foundation-language-1 | 56.0 |
| النموذج 27 | 91.04 |
| unifying-language-learning-paradigms | 29.8 |
| massive-language-models-can-be-accurately | 38.99 |
| palm-2-technical-report-1 | 69.2 |
| llama-open-and-efficient-foundation-language-1 | 47.6 |
| pythia-a-suite-for-analyzing-large-language | 36.8 |
| massive-language-models-can-be-accurately | 41.3 |
| pythia-a-suite-for-analyzing-large-language | 31.8 |
| massive-language-models-can-be-accurately | 39.85 |
| large-language-models-can-self-improve | 87.2 |
| finetuned-language-models-are-zero-shot | 63.1 |
| llama-open-and-efficient-foundation-language-1 | 52.7 |
| model-card-and-evaluations-for-claude-models | 85.7 |
| designing-effective-sparse-expert-models | 56.9 |
| llama-open-and-efficient-foundation-language-1 | 57.8 |
| bloomberggpt-a-large-language-model-for | 50.85 |
| model-card-and-evaluations-for-claude-models | 90 |
| large-language-models-can-self-improve | 88.7 |
| language-models-are-few-shot-learners | 53.2 |
| bloomberggpt-a-large-language-model-for | 48.63 |
| unifying-language-learning-paradigms | 42.9 |
| bloomberggpt-a-large-language-model-for | 45.39 |
| finetuned-language-models-are-zero-shot | 63.8 |
| galactica-a-large-language-model-for-science-1 | 51.4 |
| mistral-7b | 55.5 |
| gpt-4-technical-report-1 | 85.2 |
| bloomberggpt-a-large-language-model-for | 44.54 |
| textbooks-are-all-you-need-ii-phi-1-5 | 44.9 |