List your company

Mathematical Reasoning on AMC (avg@k, pass@k)

58.28Average @ kTeacher
8.162421.173734.18547.1963Jun 8, 2026

Evaluation results

MethodLinks
TeacherModel Scale=Teacher2026.0658.2877.11
KAT-OPDModel Scale=Qwen3-4B-Base, Rollout Length=7362026.0646.6975.9
OPDModel Scale=Qwen3-4B-Base, Rollout Length=19472026.0645.4873.49
Fixed PrefixModel Scale=Qwen3-4B-Base, Rollout Length=9872026.0645.1972.31
Random TerminationModel Scale=Qwen3-4B-Base, Rollout Length=8632026.0644.1368.68
KAT-OPDModel Scale=Qwen3-1.7B-Base, Rollout Length=4562026.0621.2456.63
Fixed PrefixModel Scale=Qwen3-1.7B-Base, Rollout Length=7302026.0620.6349.4
OPDModel Scale=Qwen3-1.7B-Base, Rollout Length=10132026.0620.3353.01
Random TerminationModel Scale=Qwen3-1.7B-Base, Rollout Length=6822026.0619.8846.99
StudentModel Scale=Qwen3-4B-Base, Evaluation Protocol=zero-shot2026.0614.9150.6
StudentModel Scale=Qwen3-1.7B-Base, Evaluation Protocol=zero-shot2026.0610.0940.96