List your company

Hallucination Detection on E2E (test)

90F1-RPoolingE
45.2856.8968.580.11Dec 28, 2023

Evaluation results

MethodLinks
PoolingEClassifier architecture=Ensembled, Training data distribution=Organic2023.1290
PoolingSLClassifier architecture=Single-layer, Training data distribution=Organic2023.1289
Expert Human2023.1279
ChatGPTPrompting strategy=Standard2023.1274
Seq-Logprob2023.1272
FActScore2023.1272
Optimized Coin (OC)Threshold optimization=Validation set tuned2023.1271
PoolingE-SYNTHClassifier architecture=Ensembled, Training data distribution=Synthetic2023.1271
PoolingSL-SYNTHClassifier architecture=Single-layer, Training data distribution=Synthetic2023.1268
ChatGPT CoTPrompting strategy=Chain-of-thought2023.1247