OSW-FT (K=2) on Qwen3-0.6B-Base achieves 0.83% Pass@1 and 6.67% Pass@16 on AIME 24, 1.04% Pass@1 and 10.00% Pass@16 on AIME 25, 8.72% Pass@1 and 41.86% Pass@16 on AMC 22, 13.04% Pass@1 and 56.52% Pass@16 on AMC 23, 5.14% Pass@1 and 28.89% Pass@16 on AMC 24, 24.20% Pass@1 on MATH-500, and 20.20% Pass@1 on GPQA-Diamond, matching or outperforming standard SFT across all evaluated benchmarks.
来源：source-paper:Table 1, Section 4.2, page 5
报告指标与观测值
pass_at_1
m-qwen3-0-6b-osw-aime24-p1
论文报告 0.83 percentage_points
实际观测 — percentage_points
pass_at_16
m-qwen3-0-6b-osw-aime24-p16
论文报告 6.67 percentage_points
实际观测 — percentage_points
pass_at_1
m-qwen3-0-6b-osw-aime25-p1
论文报告 1.04 percentage_points
实际观测 — percentage_points
pass_at_16
m-qwen3-0-6b-osw-aime25-p16
论文报告 10 percentage_points
实际观测 — percentage_points
pass_at_1
m-qwen3-0-6b-osw-amc22-p1
论文报告 8.72 percentage_points
实际观测 — percentage_points
pass_at_16
m-qwen3-0-6b-osw-amc22-p16
论文报告 41.86 percentage_points
实际观测 — percentage_points
pass_at_1
m-qwen3-0-6b-osw-amc23-p1
论文报告 13.04 percentage_points
实际观测 — percentage_points
pass_at_16
m-qwen3-0-6b-osw-amc23-p16
论文报告 56.52 percentage_points
实际观测 — percentage_points
pass_at_1
m-qwen3-0-6b-osw-amc24-p1
论文报告 5.14 percentage_points
实际观测 — percentage_points
pass_at_16
m-qwen3-0-6b-osw-amc24-p16
论文报告 28.89 percentage_points
实际观测 — percentage_points
pass_at_1
m-qwen3-0-6b-osw-math500-p1
论文报告 24.2 percentage_points
实际观测 — percentage_points
pass_at_1
m-qwen3-0-6b-osw-gpqad-p1
论文报告 20.2 percentage_points
实际观测 — percentage_points
Assessment（0）
这条 Claim 暂无已发布的不可变 Assessment。
关联运行（0）
这条 Claim 暂无关联执行。