Honesty Evaluation on FActScore v1.0
47.3ScorePromptCD
Evaluation results
| Method | Links | ||
|---|---|---|---|
| PromptCDModel=MISTRALV0.3-7B-INSTRUCT2026.02 | 47.3 | 26.7 | |
| SLEDModel=MISTRALV0.3-7B-INSTRUCT2026.02 | 46.3 | 30.5 | |
| RECITEModel=MISTRALV0.3-7B-INSTRUCT2026.02 | 46.3 | 30.5 | |
| DoLAModel=MISTRALV0.3-7B-INSTRUCT2026.02 | 46.2 | 40.5 | |
| PromptCDModel=LLAMA3-8B-INSTRUCT2026.02 | 44.5 | 63.4 | |
| VanillaModel=MISTRALV0.3-7B-INSTRUCT2026.02 | 44.5 | 22.5 | |
| DoLAModel=LLAMA3-8B-INSTRUCT2026.02 | 41.8 | 54.3 | |
| RECITEModel=LLAMA3-8B-INSTRUCT2026.02 | 37.3 | 28.6 | |
| PromptCDModel=LLAMA2-7B-CHAT2026.02 | 36.6 | 46.4 | |
| RECITEModel=LLAMA2-7B-CHAT2026.02 | 36.1 | 32.4 | |
| SLEDModel=LLAMA2-7B-CHAT2026.02 | 36 | 32.4 | |
| DoLaModel=LLAMA2-7B-CHAT2026.02 | 35.6 | 54.1 | |
| PromptCDModel=QWEN2.5-7B-INSTRUCT2026.02 | 34.3 | 29 | |
| VanillaModel=LLAMA3-8B-INSTRUCT2026.02 | 34.2 | 20.8 | |
| VanillaModel=LLAMA2-7B-CHAT2026.02 | 33.2 | 25.6 | |
| SLEDModel=QWEN2.5-7B-INSTRUCT2026.02 | 33.2 | 27.2 | |
| RECITEModel=QWEN2.5-7B-INSTRUCT2026.02 | 30.8 | 44.3 | |
| SLEDModel=LLAMA3-8B-INSTRUCT2026.02 | 30.4 | 14.6 | |
| VanillaModel=QWEN2.5-7B-INSTRUCT2026.02 | 29.6 | 26.9 | |
| DoLAModel=QWEN2.5-7B-INSTRUCT2026.02 | 26.2 | 38.1 |