Question Answering on NQ (Cohen's d)
0.9448Cohen's dDebiased-Plausibility
Evaluation Results
| Method | Links | |
|---|---|---|
| Debiased-PlausibilityCategory=Entropy-Plausibility, alpha=0.21, N=62026.05 | 0.9448 | |
| PlausibilityCategory=Entropy-Plausibility, N=62026.05 | 0.88 | |
| LLaMA 3.3 70bCategory=Uncertainty-based2026.05 | 0.3265 | |
| Retrieval ComplexityCategory=Retriever-based2026.05 | 0.2781 | |
| Debiased-PlausibilityCategory=Avg-Plausibility, alpha=0.11, N=192026.05 | 0.1869 | |
| PlausibilityCategory=Avg-Plausibility, N=192026.05 | 0.1663 | |
| LLaMA 3.1 8bCategory=Uncertainty-based2026.05 | 0.1556 | |
| PopQACategory=Popularity, N=42026.05 | 0.1535 | |
| LLaMA 3.1 8bCategory=Prompt-based2026.05 | 0.0762 | |
| LLaMA 3.3 70bCategory=Prompt-based2026.05 | 0.0307 | |
| Flesch-KincaidCategory=Readability2026.05 | -0.424 | |
| Gunning-FogCategory=Readability2026.05 | -0.5775 |