Logical Query Answering on CN15k (manually annotated)
48.9AccuracySRC
Evaluation results
| Method | Links | |
|---|---|---|
| SRCtype=symbolic inference2024.03 | 48.9 | |
| Llama3.3 70BModel Scale=70B2024.03 | 42.6 | |
| GPT-4-previewVersion=4-preview2024.03 | 37.8 | |
| Gemini-1.5-proVersion=1.5-pro2024.03 | 37.1 | |
| GPT-3.5-turboVersion=3.5-turbo2024.03 | 34.3 |