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

Prime

Benchmarks

Task NameDataset NameSOTA ResultTrend
Process-outcome alignmentPRIME Average
Outcome Score91.81
34
Process-outcome alignmentPRIME Biology
Outcome Score91.94
34
Process-outcome alignmentPRIME Chemistry
Outcome Score93.35
34
Process-outcome alignmentPRIME Physics
Outcome Score92.41
34
Process-outcome alignmentPRIME Math
Outcome Score92.44
34
Knowledge Graph RetrievalPrime (test)
H@167.8
33
Gleason Score ClassificationPRIME Synthetic Distortion (test)
Accuracy58.33
8
PI-RADS Score ClassificationPRIME Synthetic Distortion (test)
Accuracy60
8
AutoformalizationPRIME
Success Count (SC)80.13
7
PI-QUAL score assessmentPRIME
B-ACC72.85
6
DWI Distortion assessmentPRIME
Balanced Accuracy69.75
6
Gleason Score ClassificationPRIME Real Distortion (test)
Accuracy56.43
6
PI-RADS Score ClassificationPRIME Real Distortion (test)
Accuracy (ACC)54.29
6
Question AnsweringPrime (test)
Win Ratio56.5
2
Showing 14 of 14 rows