List your company

Mathematical Reasoning on AIME 2024 (Mean@32)

19.9Mean Score (k=32)Qwen3-4B-Base + Resample w/ LOPE (w/ Training Signal Shaping)
4.74728.681112.61516.5489May 7, 2026

Evaluation results

MethodLinks
Qwen3-4B-Base + Resample w/ LOPE (w/ Training Signal Shaping)Model=Qwen3-4B-Base, Optimization Method=GRPO + Resample, Prompt Strategy=LOPE, Training Signal Shaping=true2026.0519.9
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=true2026.0519.58
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=false2026.0519.01
Qwen2.5-Math-7B + GRPOModel=Qwen2.5-Math-7B, Optimization Method=GRPO2026.0517.66
Qwen2.5-Math-7B + Resample w/ Naive PromptModel=Qwen2.5-Math-7B, Optimization Method=GRPO + Resample, Prompt Strategy=Naive Prompt2026.0517.19
Qwen3-4B-Base + GRPOModel=Qwen3-4B-Base, Optimization Method=GRPO2026.0516.41
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=false2026.0515.31
Qwen3-4B-Base + Resample w/ Naive PromptModel=Qwen3-4B-Base, Optimization Method=GRPO + Resample, Prompt Strategy=Naive Prompt2026.0514.9
Qwen2.5-Math-7BModel=Qwen2.5-Math-7B, Optimization Method=Base2026.0512.9
Qwen3-4B-BaseModel=Qwen3-4B-Base, Optimization Method=Base2026.059.38
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=true2026.058.8
Qwen3-1.7B-Base + Resample w/ Naive PromptModel=Qwen3-1.7B-Base, Optimization Method=GRPO + Resample, Prompt Strategy=Naive Prompt2026.058.7
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=false2026.057.97
Qwen3-1.7B-Base + GRPOModel=Qwen3-1.7B-Base, Optimization Method=GRPO2026.056.15
Qwen3-1.7B-BaseModel=Qwen3-1.7B-Base, Optimization Method=Base2026.055.33