Response Evaluation on FAITHQA (Misinterpretation)
29Perfect ScoreClaude-3.5
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Claude-3.52025.06 | 29 | 5.92 | |
| Mistral-7B2025.06 | 23 | 4.46 | |
| Claude-32025.06 | 22 | 5.61 | |
| GPT-4o-mini2025.06 | 11 | 5.44 | |
| LLAMA3-8B2025.06 | 11 | 5.58 | |
| Mistral-7B2025.06 | 11 | 6.08 | |
| GPT-4o2025.06 | 9 | 5.73 | |
| Claude-3.52025.06 | 9 | 7.87 | |
| LLAMA3-70B2025.06 | 7 | 4.78 | |
| GPT-4o-mini2025.06 | 7 | 7.71 | |
| LLAMA3-8B2025.06 | 7 | 7.24 | |
| GPT-4o2025.06 | 6 | 7.6 | |
| Claude-32025.06 | 5 | 7.41 | |
| LLAMA3-70B2025.06 | 4 | 7.35 |