Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Common Sense Reasoning

Benchmarks

Task NameDataset NameSOTA ResultTrend
Zero-shot Common Sense ReasoningCommon Sense Reasoning
Zero-shot Accuracy72.6
137
Common Sense Reasoning7 common sense reasoning datasets
Average Performance71.72
61
Common-sense ReasoningCommon-sense Reasoning Average
Average Accuracy73.44
39
Commonsense ReasoningCommon Sense Reasoning ARC-C, ARC-E, PIQA, StoryCloze
Average Accuracy55.58
34
Common-sense reasoningCommon-sense reasoning (BoolQ, SciQ, PIQA, WinoG., ARC-C, HellaS.)
BoolQ Accuracy88
15
Common Sense ReasoningSix common sense reasoning tasks
Accuracy69.46
15
Common Sense ReasoningCommon Sense Reasoning ARC, ARC-Easy, HellaSwag, OpenBookQA, PIQA, RACE, WinoGrande
ARC Accuracy47.3
13
Common Sense ReasoningCommon Sense Reasoning (ARC, ARE, HS, OB, PI, RA, WG)
ARC Score39.3
12
Common Sense ReasoningCommon Sense Reasoning
ARC Accuracy47.9
10
Showing 9 of 9 rows