Confidence Estimation on MKQA Spanish es (test)
82AUROCCross-lingual Probe
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Cross-lingual ProbeBackbone=Qwen 3 8B, Evaluation Protocol=Zero-shot2026.05 | 82 | 53 | 19 | 20 | |
| Seq. LikelihoodBackbone=Qwen 3 8B, Evaluation Protocol=Zero-shot2026.05 | 81 | 52 | 51 | 61 | |
| Verbalized Conf.Backbone=Qwen 3 8B, Evaluation Protocol=Zero-shot2026.05 | 75 | 51 | 54 | 54 | |
| Mass-Mean ProbeBackbone=Qwen 3 8B, Evaluation Protocol=Zero-shot2026.05 | 65 | 52 | 45 | 10 | |
| P(True)Backbone=Qwen 3 8B, Evaluation Protocol=Zero-shot2026.05 | 30 | 13 | 19 | 19 |