List your company

Graduate-level Reasoning on SuperGPQA (Pass@1)

36Pass@1Rule+CER
20.424.4528.532.55Mar 11, 2026

Evaluation results

MethodLinks
Rule+CERBackbone=Qwen3-8B-Base, Training Dataset=mathematical dataset2026.0336
Rule-basedBackbone=Qwen3-8B-Base, Training Dataset=mathematical dataset2026.0335.1
CERBackbone=Qwen3-8B-Base, Training Dataset=mathematical dataset2026.0335
General-verifierBackbone=Qwen3-8B-Base, Training Dataset=mathematical dataset2026.0334.9
Exact-matchBackbone=Qwen3-8B-Base, Training Dataset=mathematical dataset2026.0333.5
VeriFreeBackbone=Qwen3-8B-Base, Training Dataset=mathematical dataset2026.0332.4
Rule-basedBackbone=Qwen3-4B-Base, Training Dataset=mathematical dataset2026.0332.2
CERBackbone=Qwen3-4B-Base, Training Dataset=mathematical dataset2026.0332.1
Rule+CERBackbone=Qwen3-4B-Base, Training Dataset=mathematical dataset2026.0331.3
VeriFreeBackbone=Qwen3-4B-Base, Training Dataset=mathematical dataset2026.0330.9
General-verifierBackbone=Qwen3-4B-Base, Training Dataset=mathematical dataset2026.0330.9
BaseBackbone=Qwen3-8B-Base, Training Dataset=mathematical dataset2026.0327
Exact-matchBackbone=Qwen3-4B-Base, Training Dataset=mathematical dataset2026.0324.2
BaseBackbone=Qwen3-4B-Base, Training Dataset=mathematical dataset2026.0321