List your company

Mathematical Reasoning on AMC 2022, 2023, 2024

61.19Accuracy@1Qwen2.5-Math-7B + Resample w/ LOPE (w/ Training Signal Shaping)
25.497234.763644.0353.2964May 7, 2026

Evaluation results

MethodLinks
Qwen2.5-Math-7B + Resample w/ LOPE (w/ Training Signal Shaping)Model=Qwen2.5-Math-7B, Optimization Method=GRPO + Resample, Prompt Strategy=LOPE, Training Signal Shaping=true2026.0561.19
Qwen3-4B-Base + Resample w/ LOPE (w/ Training Signal Shaping)Model=Qwen3-4B-Base, Optimization Method=GRPO + Resample, Prompt Strategy=LOPE, Training Signal Shaping=true2026.0558.21
Qwen3-4B-Base + Resample w/ LOPE (w/o Training Signal Shaping)Model=Qwen3-4B-Base, Optimization Method=GRPO + Resample, Prompt Strategy=LOPE, Training Signal Shaping=false2026.0552.99
Qwen2.5-Math-7B + Resample w/ Naive PromptModel=Qwen2.5-Math-7B, Optimization Method=GRPO + Resample, Prompt Strategy=Naive Prompt2026.0550
Qwen3-4B-Base + GRPOModel=Qwen3-4B-Base, Optimization Method=GRPO2026.0547.76
Qwen2.5-Math-7B + GRPOModel=Qwen2.5-Math-7B, Optimization Method=GRPO2026.0547.76
Qwen2.5-Math-7B + Resample w/ LOPE (w/o Training Signal Shaping)Model=Qwen2.5-Math-7B, Optimization Method=GRPO + Resample, Prompt Strategy=LOPE, Training Signal Shaping=false2026.0547.01
Qwen3-4B-Base + Resample w/ Naive PromptModel=Qwen3-4B-Base, Optimization Method=GRPO + Resample, Prompt Strategy=Naive Prompt2026.0545.52
Qwen2.5-Math-7BModel=Qwen2.5-Math-7B, Optimization Method=Base2026.0535.4
Qwen3-1.7B-Base + Resample w/ LOPE (w/o Training Signal Shaping)Model=Qwen3-1.7B-Base, Optimization Method=GRPO + Resample, Prompt Strategy=LOPE, Training Signal Shaping=false2026.0533.58
Qwen3-1.7B-Base + Resample w/ LOPE (w/ Training Signal Shaping)Model=Qwen3-1.7B-Base, Optimization Method=GRPO + Resample, Prompt Strategy=LOPE, Training Signal Shaping=true2026.0532.84
Qwen3-4B-BaseModel=Qwen3-4B-Base, Optimization Method=Base2026.0532.84
Qwen3-1.7B-Base + Resample w/ Naive PromptModel=Qwen3-1.7B-Base, Optimization Method=GRPO + Resample, Prompt Strategy=Naive Prompt2026.0528.36
Qwen3-1.7B-Base + GRPOModel=Qwen3-1.7B-Base, Optimization Method=GRPO2026.0527.61
Qwen3-1.7B-BaseModel=Qwen3-1.7B-Base, Optimization Method=Base2026.0526.87