Question Answering on TruthfulQA (Macro-F1)
61.08Macro-F1MATCHA
Evaluation Results
| Method | Links | |
|---|---|---|
| MATCHAThreshold=fixed midpoint threshold2026.05 | 61.08 | |
| BLEURTThreshold=fixed midpoint threshold2026.05 | 56.03 | |
| METEORThreshold=fixed midpoint threshold2026.05 | 55.2 | |
| R1-F1Threshold=fixed midpoint threshold2026.05 | 55.04 | |
| RL-F1Threshold=fixed midpoint threshold2026.05 | 51.53 | |
| R2-F1Threshold=fixed midpoint threshold2026.05 | 44.29 | |
| SimCSEThreshold=fixed midpoint threshold2026.05 | 34.65 | |
| EmbSimThreshold=fixed midpoint threshold2026.05 | 33.95 | |
| BERTScoreThreshold=fixed midpoint threshold2026.05 | 33.33 |