| Rank | Model | Hints | Score | 95% CI (±) | Rank Spread | Provider |
|---|
Confidence Intervals on Model Score Model score is a 95%-CI estimation of the % of correct answers a human annotator would have attributed to the model. Answer correctness is based on both answer and justification.
This work uses material from the Professor Layton wiki at Fandom and is licensed under the Creative Commons Attribution-Share Alike License.