Question Clarification on AmbigQA High Aleatoric Uncertainty Superset (top 20% examples)
43.87Clarification RateICE
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| ICEBackbone=GPT-42026.05 | 43.87 | -0.0633 | 6.7878 | |
| Deep EnsemblesBackbone=GPT-42026.05 | 38.31 | -0.0076 | 7.4545 | |
| Semantic EntropyBackbone=GPT-42026.05 | 34.23 | 0.033 | 6.4545 | |
| ShaQBackbone=GPT-42026.05 | 33.65 | 0.0389 | 5.5151 |