Question Answering on MuSiQue (Cohen's d)
1.0888Cohen's dDebiased-Plausibility
Evaluation Results
| Method | Links | |
|---|---|---|
| Debiased-PlausibilityCategory=Entropy-Plausibility, alpha=0.17, N=62026.05 | 1.0888 | |
| PlausibilityCategory=Entropy-Plausibility, N=92026.05 | 0.894 | |
| LLaMA 3.3 70bCategory=Uncertainty-based2026.05 | 0.4219 | |
| LLaMA 3.3 70bCategory=Prompt-based2026.05 | 0.2453 | |
| LLaMA 3.1 8bCategory=Uncertainty-based2026.05 | 0.1365 | |
| Retrieval ComplexityCategory=Retriever-based2026.05 | 0.1284 | |
| PopQACategory=Popularity, N=202026.05 | -0.0275 | |
| Debiased-PlausibilityCategory=Avg-Plausibility, alpha=0.82, N=202026.05 | -0.2242 | |
| Gunning-FogCategory=Readability2026.05 | -0.3947 | |
| PlausibilityCategory=Avg-Plausibility, N=182026.05 | -0.4173 | |
| LLaMA 3.1 8bCategory=Prompt-based2026.05 | -0.535 | |
| Flesch-KincaidCategory=Readability2026.05 | -0.543 |