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

Math Reasoning on AIME 2024 (avg@16 accuracy)

18.33Accuracy (avg@16)

GT-Reward

Updated 1mo ago

Evaluation Results

Method	Links
GT-Reward 2026.04		18.33
JURY-RL 2026.04		13.75
LLM-KD 2026.04		12.08
Majority-Voting 2026.04		11.25
CoReward 2026.04		11.25
Entropy 2026.04		11.04
GT-Reward 2026.04		10.62
LLM-as-a-Judge 2026.04		10.42
LLM-KD 2026.04		10.21
Self-Certainty 2026.04		9.79
LLM-as-a-Judge 2026.04		9.38
JURY-RL 2026.04		9.17
Majority-Voting 2026.04		8.33
JURY-RL 2026.04		7.71
CoReward 2026.04		7.71
GT-Reward 2026.04		6.88
CoReward 2026.04		6.88
Before RL 2026.04		6.88
LLM-KD 2026.04		6.67
Entropy 2026.04		6.67
LLM-as-a-Judge 2026.04		5.83
Before RL 2026.04		4.38
Self-Certainty 2026.04		3.54
Entropy 2026.04		3.54
Before RL 2026.04		3.12
Self-Certainty 2026.04		2.71
Majority-Voting 2026.04		2.08