Question Answering on QASC (Cohen's d)
0.803Cohen's dDebiased-Plausibility
Evaluation Results
| Method | Links | |
|---|---|---|
| Debiased-PlausibilityCategory=Entropy-Plausibility, alpha=0.61, N=122026.05 | 0.803 | |
| PlausibilityCategory=Entropy-Plausibility, N=102026.05 | 0.5614 | |
| Debiased-PlausibilityCategory=Avg-Plausibility, alpha=0.53, N=132026.05 | 0.4784 | |
| PlausibilityCategory=Avg-Plausibility, N=102026.05 | 0.2724 | |
| Retrieval ComplexityCategory=Retriever-based2026.05 | 0.2225 | |
| LLaMA 3.3 70bCategory=Uncertainty-based2026.05 | 0.2119 | |
| LLaMA 3.3 70bCategory=Prompt-based2026.05 | 0.2032 | |
| LLaMA 3.1 8bCategory=Uncertainty-based2026.05 | 0.1543 | |
| Flesch-KincaidCategory=Readability2026.05 | 0.1496 | |
| LLaMA 3.1 8bCategory=Prompt-based2026.05 | 0.1065 | |
| Gunning-FogCategory=Readability2026.05 | -0.0944 | |
| PopQACategory=Popularity, N=42026.05 | -0.3206 |