Question Answering on ARC Challenge (AUROC and ECE)
86.8AUROCLLM Agg.
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| LLM Agg.ID=B62026.03 | 86.8 | 0.044 | |
| Vote CountID=B12026.03 | 79.6 | 0.089 | |
| Vote EntropyID=B22026.03 | 79.4 | 0.185 | |
| Self-Consist.ID=B42026.03 | 79.4 | 0.185 | |
| DISCOUQ-EmbedID=M22026.03 | 77.9 | 0.039 | |
| Embed Centr.ID=B52026.03 | 77.8 | 0.056 | |
| DISCOUQ-LLMID=M12026.03 | 76.6 | 0.024 | |
| Verb. Conf.ID=B32026.03 | 76 | 0.042 | |
| DISCOUQ-LearnID=M32026.03 | 73.1 | 0.009 |