List your company
ResearchFormerly WizWand

Mathematical Reasoning on HLE decontaminated (Accuracy)

8.4AccuracyDeepMath-103K
2.8884.3195.757.181May 26, 2026
Updated 5mo ago

Evaluation results

MethodLinks
DeepMath-103KBackbone=Qwen3-1.7B, Sampling Strategy=Mean@42026.058.4
DAPO++Backbone=Qwen3-1.7B, Sampling Strategy=Mean@42026.056.8
DeepScaleRBackbone=Qwen3-1.7B, Sampling Strategy=Mean@42026.056.3
Qwen3-1.7B-BaseBackbone=Qwen3-1.7B, Sampling Strategy=Mean@42026.055.9
DAPO-Math-17kBackbone=Qwen3-1.7B, Sampling Strategy=Mean@42026.055.9
Qwen3-8B-BaseBackbone=Qwen3-8B, Sampling Strategy=Mean@42026.055.7
Skywork-OR1-RL-DataBackbone=Qwen3-1.7B, Sampling Strategy=Mean@42026.055.1
OpenR1-Math-220kBackbone=Qwen3-1.7B, Sampling Strategy=Mean@42026.054.7
DAPO++Backbone=Qwen3-8B, Sampling Strategy=Mean@42026.054.7
DeepScaleRBackbone=Qwen3-8B, Sampling Strategy=Mean@42026.054.5
DeepMath-103KBackbone=Qwen3-8B, Sampling Strategy=Mean@42026.054.5
DAPO-Math-17kBackbone=Qwen3-8B, Sampling Strategy=Mean@42026.054.1
OpenR1-Math-220kBackbone=Qwen3-8B, Sampling Strategy=Mean@42026.053.9
Skywork-OR1-RL-DataBackbone=Qwen3-8B, Sampling Strategy=Mean@42026.053.1