List your company

Mathematical Reasoning on AIME 24 (Mean@16, Pass@16)

18.33Mean@16ADS
2.01246.248710.48514.7213Jun 21, 2026

Evaluation results

MethodLinks
ADSBackbone=Olmo3-7B-SFT2026.0618.3354.45
DOTS+RRBackbone=Olmo3-7B-SFT2026.0615.2850
DOTS+RRBackbone=Qwen3-4B-Base2026.0614.0934.45
ADSBackbone=Qwen3-4B-Base2026.0612.533.33
GRPOBackbone=Qwen3-4B-Base2026.0612.2231.11
ADSBackbone=Qwen2.5-Math-1.5B2026.069.4432.22
DOTS+RRBackbone=Qwen2.5-Math-1.5B2026.068.7533.33
GRPOBackbone=Olmo3-7B-SFT2026.066.6733.33
BaseBackbone=Qwen3-4B-Base2026.065.8327.78
GRPOBackbone=Qwen2.5-Math-1.5B2026.065.6334.44
BaseBackbone=Olmo3-7B-SFT2026.064.3124.44
BaseBackbone=Qwen2.5-Math-1.5B2026.062.6424.45