Preference Estimation on Sports car dataset Empirical Setting II, m=50
421.27Beta Estimate AGPT-4o
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| GPT-4oPrompt=Basic2024.12 | 421.27 | 292.59 | -25.68 | 0 | |
| GPT-4oPrompt=CoT2024.12 | 304.98 | 229.8 | -28.65 | 0 | |
| GPT-4o Fine-tunedPrompt=Basic2024.12 | 201.95 | 133.85 | -36.74 | 0 | |
| GPT-4Prompt=CoT2024.12 | 173.34 | 132.71 | -38.32 | 0 | |
| GPT-4oPrompt=FS2024.12 | 153.33 | 89.9 | -31.61 | 0 | |
| GPT-3.5-Turbo-0613Prompt=CoT2024.12 | 148.81 | 119.98 | -34.33 | 0 | |
| GPT-3.5-Turbo-0125Prompt=CoT2024.12 | 129.8 | 112.47 | -35.35 | 0 | |
| GPT-3.5-Turbo-0125Prompt=Basic2024.12 | 29.7 | 5.58 | -30.22 | 0 | |
| GPT-4Prompt=Basic2024.12 | -12.43 | -16.7 | -30.63 | 0.0002 | |
| GPT-3.5-Turbo-0613Prompt=Basic2024.12 | -17.27 | -34.2 | -37.81 | 0.105 |