List your company

Confidence Estimation on MKQA Russian / ru (test)

80AUROCCross-lingual Probe
41.5251.5161.571.49May 29, 2026

Evaluation results

MethodLinks
Cross-lingual ProbeBackbone=Qwen 3 8B, Evaluation Protocol=Zero-shot2026.0580431516
Seq. LikelihoodBackbone=Qwen 3 8B, Evaluation Protocol=Zero-shot2026.0579354962
Mass-Mean ProbeBackbone=Qwen 3 8B, Evaluation Protocol=Zero-shot2026.0573514012
Verbalized Conf.Backbone=Qwen 3 8B, Evaluation Protocol=Zero-shot2026.0572365556
P(True)Backbone=Qwen 3 8B, Evaluation Protocol=Zero-shot2026.0543121212