List your company

Confidence Estimation on MKQA Japanese ja (test)

83AUROCMass-Mean Probe
59.0865.2971.577.71May 29, 2026

Evaluation results

MethodLinks
Mass-Mean ProbeBackbone=Qwen 3 8B, Evaluation Protocol=Zero-shot2026.058357229
Seq. LikelihoodBackbone=Qwen 3 8B, Evaluation Protocol=Zero-shot2026.0581414356
Verbalized Conf.Backbone=Qwen 3 8B, Evaluation Protocol=Zero-shot2026.0567376671
Cross-lingual ProbeBackbone=Qwen 3 8B, Evaluation Protocol=Zero-shot2026.0564231616
P(True)Backbone=Qwen 3 8B, Evaluation Protocol=Zero-shot2026.0560161313