Multiple Choice Question Answering on Aggregate MCQA (average across datasets)
68.4PrecisionSecond Guess
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Second Guess2026.05 | 68.4 | 20.12 | 29.44 | |
| Entropy ThresholdingPrompting Strategy=Augmented2026.05 | 63.49 | 29.88 | 34.69 | |
| Entropy ThresholdingPrompting Strategy=Original2026.05 | 63.3 | 32 | 36.56 | |
| Self-Evaluation2026.05 | 62.1 | 32.56 | 32.56 | |
| AugmentedPrompting Strategy=Augmented2026.05 | 61.04 | 38 | 38 | |
| OriginalPrompting Strategy=Original2026.05 | 59.75 | 40.25 | 40.25 |