Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
STEM Reasoning on GaokaoCloze
Loading...
25.9
Accuracy
AGPO
18.516
20.433
22.35
24.267
May 20, 2026
Accuracy
Updated 13d ago
Evaluation Results
Method
Method
Links
Accuracy
AGPO
clipping=adaptive, ATS...
2026.05
25.9
AGPO
clipping=adaptive, ATS...
2026.05
21.2
GRPO
ATS=true
2026.05
20.5
GRPO
clipping=fixed ε
2026.05
20.3
Adaptive-KL PPO
2026.05
19.6
PPO
2026.05
18.9
DPO
2026.05
18.8
Feedback
Search any
task
Search any
task
