Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Mathematical Reasoning on Olympiad (Accuracy, Avg)
Loading...
47
Accuracy (Olympiad)
GRPO + coupled rhythm credit
43.152
44.151
45.15
46.149
Oct 15, 2025
Accuracy (Olympiad)
Average Score (Olympiad)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy (Olympiad)
Average Score (Olympiad)
GRPO + coupled rhythm credit
Base Model=Qwen3-8B-Ba...
2025.10
47
43.2
GRPO + Reweight+Lopti
Base Model=Qwen3-8B-Ba...
2025.10
46.5
41.1
GRPO + global-anchor credit
Base Model=Qwen3-8B-Ba...
2025.10
46.1
42.1
GRPO + CAPO
Base Model=Qwen3-8B-Ba...
2025.10
45.9
41.4
GRPO + local-chunk credit
Base Model=Qwen3-8B-Ba...
2025.10
45.9
41.3
GRPO + high-entropy credit
Base Model=Qwen3-8B-Ba...
2025.10
45.6
40.1
GRPO + high-entropy selection
Base Model=Qwen3-8B-Ba...
2025.10
45.6
40.9
GRPO + ThinkPRM-1.5B
Base Model=Qwen3-8B-Ba...
2025.10
45.4
41
GRPO + gradient-based credit
Base Model=Qwen3-8B-Ba...
2025.10
45.4
40
GRPO + token-correlation credit
Base Model=Qwen3-8B-Ba...
2025.10
45.3
37.4
GRPO + path-aggregation credit
Base Model=Qwen3-8B-Ba...
2025.10
45.1
40
GRPO + AsyPPO
Base Model=Qwen3-8B-Ba...
2025.10
44.7
40
GRPO
Base Model=Qwen3-8B-Ba...
2025.10
44.2
39.4
GRPO + random credit
Base Model=Qwen3-8B-Ba...
2025.10
43.3
39.6
Feedback
Search any
task
Search any
task