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

Selective Prediction on GSM8K

90.4PRR

Evo-Anth-1

85.61686.85888.189.342Apr 3, 2026
Updated 12d ago

Evaluation Results

MethodLinks
2026.04
90.4
2026.04
90.4
2026.04
90.3
2026.04
90.2
2026.04
90.2
2026.04
90.1
2026.04
90.1
2026.04
89.9
2026.04
89.9
89.9
2026.04
89.8
2026.04
89.1
2026.04
89
2026.04
88.5
2026.04
88.4
88.4
87.5
2026.04
86.7
2026.04
86
2026.04
85.8