List your company

Mathematical Reasoning on AIME (Exact Match Accuracy)

12.31Exact Match AccuracyExplorer-only, successful seeds (diagnostic)
1.42124.24817.0759.9019May 8, 2026

Evaluation results

MethodLinks
Explorer-only, successful seeds (diagnostic)Backbone=Qwen3-8B-Base2026.0512.31
Entropy reweightingBackbone=Qwen3-8B-Base, Layer=Mid (Layer 20)2026.059.75
Entropy reweightingBackbone=Qwen3-8B-Base, Layer=Deep (Layer 31)2026.059.14
Random-20%Backbone=Qwen3-8B-Base2026.058.12
Entropy reweightingBackbone=Qwen3-8B-Base, Layer=Shallow (Layer 8)2026.057.37
Full-token DAPOBackbone=Qwen3-8B-Base2026.055.83
Anchor-onlyBackbone=Qwen3-8B-Base2026.055.6
Explorer-only, all seedsBackbone=Qwen3-8B-Base2026.054.62
High Prediction Entropy-onlyBackbone=Qwen3-8B-Base2026.051.84