List your company
ResearchFormerly WizWand

Mathematical Reasoning on AMO decontaminated (Accuracy)

3.9AccuracyDAPO++
0.8841.6672.453.233May 26, 2026
Updated 5mo ago

Evaluation results

MethodLinks
DAPO++Backbone=Qwen3-8B, Sampling Strategy=Mean@322026.053.9
DeepMath-103KBackbone=Qwen3-8B, Sampling Strategy=Mean@322026.053.7
OpenR1-Math-220kBackbone=Qwen3-8B, Sampling Strategy=Mean@322026.053.3
DeepMath-103KBackbone=Qwen3-1.7B, Sampling Strategy=Mean@322026.053.2
DeepScaleRBackbone=Qwen3-8B, Sampling Strategy=Mean@322026.052.9
Skywork-OR1-RL-DataBackbone=Qwen3-8B, Sampling Strategy=Mean@322026.052.8
DAPO-Math-17kBackbone=Qwen3-8B, Sampling Strategy=Mean@322026.052.4
Skywork-OR1-RL-DataBackbone=Qwen3-1.7B, Sampling Strategy=Mean@322026.052
Qwen3-8B-BaseBackbone=Qwen3-8B, Sampling Strategy=Mean@322026.051.8
DAPO++Backbone=Qwen3-1.7B, Sampling Strategy=Mean@322026.051.7
OpenR1-Math-220kBackbone=Qwen3-1.7B, Sampling Strategy=Mean@322026.051.4
Qwen3-1.7B-BaseBackbone=Qwen3-1.7B, Sampling Strategy=Mean@322026.051.3
DAPO-Math-17kBackbone=Qwen3-1.7B, Sampling Strategy=Mean@322026.051.3
DeepScaleRBackbone=Qwen3-1.7B, Sampling Strategy=Mean@322026.051