Confidence Estimation on MKQA Japanese ja (test)
83AUROCMass-Mean Probe
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Mass-Mean ProbeBackbone=Qwen 3 8B, Evaluation Protocol=Zero-shot2026.05 | 83 | 57 | 22 | 9 | |
| Seq. LikelihoodBackbone=Qwen 3 8B, Evaluation Protocol=Zero-shot2026.05 | 81 | 41 | 43 | 56 | |
| Verbalized Conf.Backbone=Qwen 3 8B, Evaluation Protocol=Zero-shot2026.05 | 67 | 37 | 66 | 71 | |
| Cross-lingual ProbeBackbone=Qwen 3 8B, Evaluation Protocol=Zero-shot2026.05 | 64 | 23 | 16 | 16 | |
| P(True)Backbone=Qwen 3 8B, Evaluation Protocol=Zero-shot2026.05 | 60 | 16 | 13 | 13 |