Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Mathematical Reasoning on AIME 24 (Accuracy Avg@8)
Loading...
82.1
Accuracy (Avg@8)
CERL full
71.596
74.323
77.05
79.777
May 22, 2026
Accuracy (Avg@8)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy (Avg@8)
CERL full
Training=SFT+RL
2026.05
82.1
Correctness-only RL
Training=RL
2026.05
79.6
Halo
Training=SFT+RL
2026.05
79.5
Long-answer SFT
Training=SFT
2026.05
78.4
Fixed-ratio erasure + free answer
Training=None
2026.05
76.4
Base Qwen3-8B
Training=N/A
2026.05
75.9
Length-penalty RL
Training=RL
2026.05
75.1
TokenSkip
Training=SFT
2026.05
72
Feedback
Search any
task
Search any
task