Visual Reasoning on Geometric Shapes
95.2AccuracyRoT
Evaluation Results
| Method | Links | |
|---|---|---|
| RoTBackbone=GPT-4, Shots=2-shot2024.10 | 95.2 | |
| RoTBackbone=GPT-4, Shots=1-shot2024.10 | 94.8 | |
| BOTBackbone=GPT-4, Shots=2-shot2024.10 | 90.7 | |
| BOTBackbone=GPT-4, Shots=1-shot2024.10 | 90.4 | |
| RoTBackbone=GPT-3.5-turbo, Shots=2-shot2024.10 | 88.7 | |
| RoTBackbone=GPT-3.5-turbo, Shots=1-shot2024.10 | 88.4 | |
| BOTBackbone=GPT-3.5-turbo, Shots=2-shot2024.10 | 84.3 | |
| BOTBackbone=GPT-3.5-turbo, Shots=1-shot2024.10 | 83.1 | |
| Meta-promptingBackbone=GPT-4, Shots=2-shot2024.10 | 78.2 | |
| Meta-promptingBackbone=GPT-4, Shots=1-shot2024.10 | 76.7 | |
| Meta-promptingBackbone=GPT-3.5-turbo, Shots=2-shot2024.10 | 71 | |
| Meta-promptingBackbone=GPT-3.5-turbo, Shots=1-shot2024.10 | 69.5 | |
| COTBackbone=GPT-4, Shots=2-shot2024.10 | 60.9 | |
| TOTBackbone=GPT-4, Shots=2-shot2024.10 | 60.1 | |
| GOTBackbone=GPT-3.5-turbo, Shots=2-shot2024.10 | 58.9 | |
| GOTBackbone=GPT-3.5-turbo, Shots=1-shot2024.10 | 58.6 | |
| TOTBackbone=GPT-4, Shots=1-shot2024.10 | 58.6 | |
| TOTBackbone=GPT-3.5-turbo, Shots=2-shot2024.10 | 58.5 | |
| GOTBackbone=GPT-4, Shots=2-shot2024.10 | 57.7 | |
| COTBackbone=GPT-4, Shots=1-shot2024.10 | 57.4 | |
| TOTBackbone=GPT-3.5-turbo, Shots=1-shot2024.10 | 56.2 | |
| GOTBackbone=GPT-4, Shots=1-shot2024.10 | 56.2 | |
| PROMPTEDPrompting Technique=PROMPTED2023.10 | 55.2 | |
| Zero-ShotPrompting Technique=Zero-Shot2023.10 | 54.4 | |
| Zero-Shot CoTPrompting Technique=Zero-Shot Chain-of-Thought2023.10 | 54.4 | |
| COTBackbone=GPT-3.5-turbo, Shots=2-shot2024.10 | 53.2 | |
| Output RefinementPrompting Technique=Output Refinement2023.10 | 52.8 | |
| COTBackbone=GPT-3.5-turbo, Shots=1-shot2024.10 | 50.2 |