List your company

Predictive Validity Verification on Downstream Tasks (KOLD, HateXplain, etc.)

0.3156Average CorrelationDOVE
-0.228632-0.0873410.053950.195241Mar 16, 2026

Evaluation results

MethodLinks
DOVEEvaluation Mode=Average across 12 LLMs2026.030.3156
CDEvalEvaluation Mode=Average across 12 LLMs2026.030.2356
WVSEvaluation Mode=Average across 12 LLMs2026.030.162
NormAdEvaluation Mode=Average across 12 LLMs2026.030.009
GOQAEvaluation Mode=Average across 12 LLMs2026.03-0.1305
NaVABEvaluation Mode=Average across 12 LLMs2026.03-0.2077