List your company

Mathematical Reasoning on MATH 500 (pass@8, avg@8)

89.38Average Score (k=8)DAPO w/ PR-FKL
59.79267.473575.15582.8365May 30, 2026

Evaluation results

MethodLinks
DAPO w/ PR-FKLModel Backbone=Qwen3-8B-Base2026.0589.3896.2
DAPO w/ RRModel Backbone=Qwen3-8B-Base2026.0589.3396.8
DAPO w/ PR-RKLModel Backbone=Qwen3-8B-Base2026.0588.5895.6
DAPOModel Backbone=Qwen3-8B-Base2026.0587.6596
DAPO w/ PR-FKLModel Backbone=Qwen3-4B-Base2026.0586.4894.4
DAPO w/ PR-RKLModel Backbone=Qwen3-4B-Base2026.0586.294.2
DAPOModel Backbone=Qwen3-4B-Base2026.0586.194.2
DAPO w/ RRModel Backbone=Qwen3-4B-Base2026.0585.9895
BaseModel Backbone=Qwen3-4B-Base2026.0562.3889
BaseModel Backbone=Qwen3-8B-Base2026.0560.9389.8