List your company

Mathematical Reasoning on MATH500 (Test accuracy (%))

0.378Test Accuracy (%)d1
0.317680.333340.3490.36466May 12, 2026

Evaluation results

MethodLinks
d1Generation Length=5122026.050.378
ESPOGeneration Length=5122026.050.376
StableDRLGeneration Length=5122026.050.376
wd1Generation Length=5122026.050.374
Diffu-GRPOGeneration Length=5122026.050.368
StableDRLGeneration Length=2562026.050.354
SPGGeneration Length=5122026.050.352
GDPOGeneration Length=5122026.050.35
ESPOGeneration Length=2562026.050.35
SPGGeneration Length=2562026.050.348
LLaDA-8B-InstructGeneration Length=5122026.050.346
SFTGeneration Length=5122026.050.344
wd1Generation Length=2562026.050.342
AdaBlock-dLLMGeneration Length=5122026.050.336
Diffu-GRPOGeneration Length=2562026.050.336
d1Generation Length=2562026.050.334
AdaBlock-dLLMGeneration Length=2562026.050.324
SFTGeneration Length=2562026.050.322
GDPOGeneration Length=2562026.050.322
LLaDA-8B-InstructGeneration Length=2562026.050.32