Hallucination Detection on CDM (test)
75F1 ScorePoolingE
Evaluation Results
| Method | Links | |
|---|---|---|
| PoolingEClassifier architecture=Ensembled, Training data distribution=Organic2023.12 | 75 | |
| PoolingSLClassifier architecture=Single-layer, Training data distribution=Organic2023.12 | 73 | |
| SummaCzsModel variant=Zero-shot2023.12 | 68 | |
| SummaCConvModel variant=Convolutional2023.12 | 67 | |
| PoolingE-SYNTHClassifier architecture=Ensembled, Training data distribution=Synthetic2023.12 | 66 | |
| PoolingSL-SYNTHClassifier architecture=Single-layer, Training data distribution=Synthetic2023.12 | 65 | |
| Seq-Logprob2023.12 | 64 | |
| ChatGPTPrompting strategy=Standard2023.12 | 63 | |
| QuestEval2023.12 | 61 | |
| FActScore2023.12 | 61 | |
| Expert Human2023.12 | 55 | |
| DAE2023.12 | 53 | |
| ChatGPT CoTPrompting strategy=Chain-of-thought2023.12 | 53 | |
| FactCC2023.12 | 51 | |
| FEQA2023.12 | 44 | |
| Optimized Coin (OC)Threshold optimization=Validation set tuned2023.12 | 43 |