Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Mathematical Reasoning on AMC 2023 (test) (pass@1)
Loading...
65.4
Pass@1 Rate
GRPO + coupled rhythm credit
54.688
57.469
60.25
63.031
Oct 15, 2025
Pass@1 Rate
Updated 1mo ago
Evaluation Results
Method
Method
Links
Pass@1 Rate
GRPO + coupled rhythm credit
Backbone=Qwen3-8B-Base...
2025.10
65.4
GRPO + global-anchor credit
Backbone=Qwen3-8B-Base...
2025.10
63.1
GRPO + local-chunk credit
Backbone=Qwen3-8B-Base...
2025.10
62.9
GRPO + global-anchor credit
Backbone=Qwen3-4B-Base...
2025.10
61.1
GRPO + high-entropy credit
Backbone=Qwen3-8B-Base...
2025.10
60.4
GRPO + random credit
Backbone=Qwen3-8B-Base...
2025.10
60.1
GRPO + coupled rhythm credit
Backbone=Qwen3-4B-Base...
2025.10
59.3
GRPO + local-chunk credit
Backbone=Qwen3-4B-Base...
2025.10
59.2
GRPO
Backbone=Qwen3-8B-Base...
2025.10
59.1
GRPO + local-chunk credit
Backbone=Qwen3-4B-Base...
2025.10
58.4
GRPO + high-entropy credit
Backbone=Qwen3-4B-Base...
2025.10
57.8
GRPO + global-anchor credit
Backbone=Qwen3-4B-Base...
2025.10
57.6
GRPO
Backbone=Qwen3-4B-Base...
2025.10
57.6
GRPO + coupled rhythm credit
Backbone=Qwen3-4B-Base...
2025.10
57.4
GRPO + random credit
Backbone=Qwen3-4B-Base...
2025.10
57.1
GRPO + high-entropy credit
Backbone=Qwen3-4B-Base...
2025.10
55.5
GRPO + random credit
Backbone=Qwen3-4B-Base...
2025.10
55.2
GRPO
Backbone=Qwen3-4B-Base...
2025.10
55.1
Feedback
Search any
task
Search any
task