List your company
ResearchFormerly WizWand

Mathematical Reasoning on MinervaMath (held-out)

49.7AccuracyAntiSD
28.27633.83839.444.962May 12, 2026
Updated 5mo ago

Evaluation results

MethodLinks
AntiSDBackbone=Qwen3-30B-A3B, Training Condition=+AntiSD, Speedup=2.9×, Peak-mean step=140, Sampling strategy=avg@42026.0549.7
AntiSDBackbone=Qwen3-8B, Training Condition=+AntiSD, Speedup=5.0×, Peak-mean step=180, Sampling strategy=avg@42026.0548.5
GRPOBackbone=Qwen3-30B-A3B, Training Condition=+GRPO, Speedup=1.0×, Peak-mean step=150, Sampling strategy=avg@42026.0547.1
AntiSDBackbone=Qwen3-4B-IT-2507, Training Condition=+AntiSD, Speedup=10.0×, Peak-mean step=100, Sampling strategy=avg@42026.0546.4
GRPOBackbone=Olmo3-7B-TK, Training Condition=+GRPO, Speedup=1.0×, Peak-mean step=80, Sampling strategy=avg@42026.0546.4
AntiSDBackbone=Olmo3-7B-TK, Training Condition=+AntiSD, Speedup=2.0×, Peak-mean step=40, Sampling strategy=avg@42026.0545.8
Olmo3-7B-TKBackbone=Olmo3-7B-TK, Training Condition=Base, Sampling strategy=avg@42026.0545.2
GRPOBackbone=Qwen3-8B, Training Condition=+GRPO, Speedup=1.0×, Peak-mean step=200, Sampling strategy=avg@42026.0545.1
SDBackbone=Olmo3-7B-TK, Training Condition=+SD, Peak-mean step=40, Sampling strategy=avg@42026.0544.8
SDBackbone=Qwen3-30B-A3B, Training Condition=+SD, Peak-mean step=60, Sampling strategy=avg@42026.0544.1
Qwen3-30B-A3BBackbone=Qwen3-30B-A3B, Training Condition=Base, Sampling strategy=avg@42026.0543.5
Qwen3-4B-IT-2507Backbone=Qwen3-4B-IT-2507, Training Condition=Base, Sampling strategy=avg@42026.0543.4
SDBackbone=Qwen3-4B-IT-2507, Training Condition=+SD, Peak-mean step=10, Sampling strategy=avg@42026.0543
AntiSDBackbone=Olmo3-7B-IT, Training Condition=+AntiSD, Speedup=9.5×, Peak-mean step=200, Sampling strategy=avg@42026.0542.4
SDBackbone=Qwen3-8B, Training Condition=+SD, Peak-mean step=200, Sampling strategy=avg@42026.0540.7
Qwen3-8BBackbone=Qwen3-8B, Training Condition=Base, Sampling strategy=avg@42026.0539
Olmo3-7B-ITBackbone=Olmo3-7B-IT, Training Condition=Base, Sampling strategy=avg@42026.0538.7
SDBackbone=Olmo3-7B-IT, Training Condition=+SD, Peak-mean step=10, Sampling strategy=avg@42026.0538.5
GRPOBackbone=Qwen3-4B-IT-2507, Training Condition=+GRPO, Speedup=1.0×, Peak-mean step=200, Sampling strategy=avg@42026.0533.2
GRPOBackbone=Olmo3-7B-IT, Training Condition=+GRPO, Speedup=1.0×, Peak-mean step=190, Sampling strategy=avg@42026.0529.1