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

Mathematical Reasoning on MATH500 (Test accuracy (%))

0.378Test Accuracy (%)

d1

Updated 21d ago

Evaluation Results

Method	Links
d1 2026.05		0.378
ESPO 2026.05		0.376
StableDRL 2026.05		0.376
wd1 2026.05		0.374
Diffu-GRPO 2026.05		0.368
StableDRL 2026.05		0.354
SPG 2026.05		0.352
GDPO 2026.05		0.35
ESPO 2026.05		0.35
SPG 2026.05		0.348
LLaDA-8B-Instruct 2026.05		0.346
SFT 2026.05		0.344
wd1 2026.05		0.342
AdaBlock-dLLM 2026.05		0.336
Diffu-GRPO 2026.05		0.336
d1 2026.05		0.334
AdaBlock-dLLM 2026.05		0.324
SFT 2026.05		0.322
GDPO 2026.05		0.322
LLaDA-8B-Instruct 2026.05		0.32