List your company

Question Answering on NQ (Cohen's d)

0.9448Cohen's dDebiased-Plausibility
-0.638392-0.2273710.183650.594671May 12, 2026

Evaluation results

MethodLinks
Debiased-PlausibilityCategory=Entropy-Plausibility, alpha=0.21, N=62026.050.9448
PlausibilityCategory=Entropy-Plausibility, N=62026.050.88
LLaMA 3.3 70bCategory=Uncertainty-based2026.050.3265
Retrieval ComplexityCategory=Retriever-based2026.050.2781
Debiased-PlausibilityCategory=Avg-Plausibility, alpha=0.11, N=192026.050.1869
PlausibilityCategory=Avg-Plausibility, N=192026.050.1663
LLaMA 3.1 8bCategory=Uncertainty-based2026.050.1556
PopQACategory=Popularity, N=42026.050.1535
LLaMA 3.1 8bCategory=Prompt-based2026.050.0762
LLaMA 3.3 70bCategory=Prompt-based2026.050.0307
Flesch-KincaidCategory=Readability2026.05-0.424
Gunning-FogCategory=Readability2026.05-0.5775