Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Mathematical Reasoning on AIME 2024 (Accuracy@8)
Loading...
87
Accuracy (Avg@8)
CERL full
75.872
78.761
81.65
84.539
May 22, 2026
Accuracy (Avg@8)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy (Avg@8)
CERL full
Training=SFT+RL
2026.05
87
Fixed-ratio erasure + free answer
Training=None
2026.05
83.7
Halo
Training=SFT+RL
2026.05
83
Base Qwen3-32B
Training=N/A
2026.05
81.5
Length-penalty RL
Training=RL
2026.05
78.8
TokenSkip†
Training=SFT
2026.05
76.3
Feedback
Search any
task
Search any
task