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

Mathematical Reasoning on MATH 500 (Speedup, Length, Cost Share)

3.14Speedup

SpecBlock+adapt

0.47761.16881.862.5512May 8, 2026
Updated 23d ago

Evaluation Results

MethodLinks
2026.05
3.144.2315
2026.05
3.14.2418
2026.05
3.074.0316
2026.05
2.673.7918
2026.05
2.665.9331
2026.05
2.643.8319
2026.05
2.613.917
2026.05
2.63.6314
2026.05
2.573.9219
2026.05
2.553.7313
2026.05
2.553.6413
2026.05
2.533.7114
2026.05
2.513.7215
2026.05
2.513.6714
2026.05
2.483.5311
2026.05
2.483.4811
2026.05
2.33.7426
2026.05
2.184.2829
2026.05
2.024.0625
2026.05
1.944.4629
2026.05
1.892.648
2026.05
1.824.2724
2026.05
1.783.3115
2026.05
1.773.3817
2026.05
1.743.2414
2026.05
1.532.36
2026.05
1.324.3130
2026.05
1.172.2938
2026.05
1.153.0127
2026.05
0.972.189
2026.05
0.831.97
2026.05
0.581.8740