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

PIQA

Benchmarks

Task NameDataset NameSOTA ResultTrend
Commonsense ReasoningPIQA
Accuracy94.9
757
Physical Commonsense ReasoningPIQA
Accuracy94.9
696
Question AnsweringPIQA
Accuracy86.5
505
Physical Interaction Question AnsweringPIQA
Accuracy94.9
415
Commonsense ReasoningPIQA
Accuracy89.99
213
ReasoningPIQA
Accuracy96.5
164
Physical Commonsense ReasoningPIQA (val)
Accuracy83
118
Common Sense ReasoningPIQA
Accuracy91.89
100
Physical Commonsense ReasoningPIQA
Accuracy (PIQA)81.5
99
Physical ReasoningPIQA
Accuracy82.5
90
Physical Commonsense ReasoningPIQA
Accuracy85.91
78
Commonsense reasoningPIQA 1.0 (test)
Accuracy82.21
64
Multiple Choice Question AnsweringPIQA
Accuracy80.5
63
Zero-shot ReasoningPIQA
PIQA Zero-shot Accuracy80.9
62
Physical Commonsense ReasoningPIQA (test)
Accuracy90.7
59
Commonsense ReasoningPIQA (test)
Accuracy90.1
57
Physical Commonsense ReasoningPIQA
Accuracy7,497
56
Physical Commonsense ReasoningPIQA
Accuracy82.54
45
Physical Commonsense ReasoningPiQA
Accuracy76.56
45
Commonsense ReasoningPIQA
Normalized Accuracy85.47
41
Question AnsweringPIQA (test)
Accuracy85
40
Question AnsweringPiQA
Accuracy81.77
36
Physical ReasoningPIQA
Accuracy91.3
34
Zero-shot AccuracyPIQA
Zero-shot PIQA Accuracy81.5
30
Inactive Attention Head IdentificationPIQA
Percentage of Heads Zeroed31.3
28
Showing 25 of 89 rows