Logical Reasoning on ReClor (val)
84.4Accuracy (ReClor val)GPT-4-0125-preview
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-4-0125-previewCategory=LLM2024.05 | 84.4 | |
| PathReasoner2024.05 | 70.4 | |
| MERITCategory=Sequence, Extra data=true2024.05 | 69.4 | |
| LogiformerCategory=Graph2024.05 | 68.4 | |
| FocalReasonerCategory=Graph2024.05 | 66.8 | |
| LReasonerCategory=Sequence2024.05 | 66.2 | |
| DAGNCategory=Graph2024.05 | 65.8 | |
| AdaLoGNCategory=Graph2024.05 | 65.2 | |
| ROBERTa-LargeCategory=Sequence2024.05 | 62.6 | |
| XLNet-LargeCategory=Sequence2024.05 | 62 | |
| GPT-3.5-turboCategory=LLM2024.05 | 58.8 | |
| PaLMv2Category=LLM, Extra data=true2024.05 | 56 | |
| BERT-LargeCategory=Sequence2024.05 | 53.8 | |
| text-davinci-003Category=LLM2024.05 | 53 | |
| Random2024.05 | 25 |