Fact Retention on key_facts
100Original ScoreGPT-4.1
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| GPT-4.1n=4,0812026.05 | 100 | 99.1 | 99 | 0.9 | 1 | 58.5 | |
| GPT-4o-minin=4,0812026.05 | 99.1 | 95.7 | 94.6 | 3.4 | 4.5 | 58.5 | |
| GPT-4.1-nanon=4,0812026.05 | 98 | 95 | 94.9 | 3 | 3.1 | 58.5 |
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| GPT-4.1n=4,0812026.05 | 100 | 99.1 | 99 | 0.9 | 1 | 58.5 | |
| GPT-4o-minin=4,0812026.05 | 99.1 | 95.7 | 94.6 | 3.4 | 4.5 | 58.5 | |
| GPT-4.1-nanon=4,0812026.05 | 98 | 95 | 94.9 | 3 | 3.1 | 58.5 |