List your company

Question Answering on TruthfulQA (AUROC and ECE Metrics)

0.871AUROCDISCOUQ-Learn
0.589160.662330.73550.80867Mar 21, 2026Mar 30, 2026Apr 9, 2026Apr 19, 2026Apr 29, 2026May 9, 2026May 19, 2026

Evaluation results

MethodLinks
DISCOUQ-LearnID=M32026.030.8710.074
DISCOUQ-LLMID=M12026.030.8670.062
Embed Centr.ID=B52026.030.8070.155
LLM Agg.ID=B62026.030.8050.102
Vote CountID=B12026.030.7980.037
Vote EntropyID=B22026.030.7980.096
Self-Consist.ID=B42026.030.7980.096
DISCOUQ-EmbedID=M22026.030.7980.039
Verb. Conf.ID=B32026.030.7330.122
RALCSignal=Sem. Uncertainty2026.050.664—
RALCSignal=Token Probability2026.050.63—
Hedged QA2026.050.627—
Original Response2026.050.61—
RALCSignal=Ling. Confidence2026.050.6—