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

Mathematical Reasoning on AIME 24 (Pass@1, TC)

26.7Pass@1 Score

EAPO

Updated 25d ago

Evaluation Results

Method	Links
EAPO 2026.06		26.7	1
DAPO 2026.06		16.7	-
DAPO (WITH FORKING TOKENS) 2026.06		16.7	-
S-GRPO 2026.06		16.7	-
PS-PPO (OPTIMIZED, B=128) 2026.06		16.7	-
Reinforce++ 2026.06		13.3	1.08
GRPO 2026.06		13.3	-
DR.GRPO 2026.06		13.3	-
PS-PPO (UNIFORM) 2026.06		13.3	-
PS-PPO (HEURISTIC) 2026.06		13.3	-
GRPO 2026.06		10	1.02
DAPO 2026.06		10	-
DAPO (WITH FORKING TOKENS) 2026.06		10	-
S-GRPO 2026.06		10	-
PS-PPO (OPTIMIZED, B=128) 2026.06		10	-
RLOO 2026.06		10	-
FIXED LENGTH (L=512) 2026.06		10	-
PS-PPO (TIME-PRIOR) 2026.06		10	-
GRPO 2026.06		6.6	-
DR.GRPO 2026.06		6.6	-
RLOO 2026.06		6.6	-
FIXED LENGTH (L=512) 2026.06		6.6	-
PS-PPO (UNIFORM) 2026.06		6.6	-
PS-PPO (TIME-PRIOR) 2026.06		6.6	-
PS-PPO (HEURISTIC) 2026.06		6.6	-
BASE 2026.06		6.6	-
TIR 2026.06		3.3	0.52
BASE 2026.06		3.3	-
Base 2026.06		0	-