Commonsense Question Answering on RiddleSense
67.3AccuracySCORE (GPT-4 Verifier)
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| SCORE (GPT-4 Verifier)Base LM=LLaMA-2-13B-chat, Verifier=gpt-4, Refiner=SCORE (fine-tuned)2024.04 | 67.3 | 87.7 | |
| SCORE (GPT-4 Verifier)Base LM=Gemma-7B-it, Verifier=gpt-4, Refiner=SCORE (fine-tuned)2024.04 | 64.9 | 85.6 | |
| Few-shot promptingBase LM=LLaMA-2-13B-chat2024.04 | 57.6 | — | |
| Few-shot promptingBase LM=Gemma-7B-it2024.04 | 57.1 | — |