Online Self-Weighted Fine-Tuning
Not assessedPlan blockedFindingclaim-main-table1-qwen3-4b
OSW-FT (K=2) on Qwen3-4B-Base achieves 13.33% Pass@1 and 36.67% Pass@16 on AIME 24, 16.25% Pass@1 and 36.67% Pass@16 on AIME 25, 41.42% Pass@1 and 74.42% Pass@16 on AMC 22, 45.11% Pass@1 and 78.26% Pass@16 on AMC 23, 35.00% Pass@1 and 75.56% Pass@16 on AMC 24, 79.00% Pass@1 on MATH-500, and 32.32% Pass@1 on GPQA-Diamond.
Source: source-paper:Table 1, Section 4.2, page 5
Reported and observed measurements
pass_at_1
m-qwen3-4b-osw-aime24-p1
Reported 13.33 percentage_points
Observed — percentage_points
pass_at_16
m-qwen3-4b-osw-aime24-p16
Reported 36.67 percentage_points
Observed — percentage_points
pass_at_1
m-qwen3-4b-osw-aime25-p1
Reported 16.25 percentage_points
Observed — percentage_points
pass_at_16
m-qwen3-4b-osw-aime25-p16
Reported 36.67 percentage_points
Observed — percentage_points
pass_at_1
m-qwen3-4b-osw-math500-p1
Reported 79 percentage_points
Observed — percentage_points
pass_at_1
m-qwen3-4b-osw-gpqad-p1
Reported 32.32 percentage_points
Observed — percentage_points
Assessments (0)
No immutable Assessment has been published for this Claim yet.
Runs (0)
No execution has been linked to this Claim yet.