List your company

Tool-Integrated Reasoning on Olympiad Bench

52.75Test AccuracySeq-ALP
48.205249.385150.56551.7449Mar 19, 2026

Evaluation results

MethodLinks
Seq-ALPtemperature=1.0, sampling=average@32, multi-turn=true2026.0352.75
Token-ALPtemperature=1.0, sampling=average@32, multi-turn=true2026.0351.55
Seq-Bypasstemperature=1.0, sampling=average@32, multi-turn=true2026.0351.08
Token-MIStemperature=1.0, sampling=average@32, multi-turn=true2026.0350.65
GRPOtemperature=1.0, sampling=average@32, multi-turn=true2026.0349.59
Seq-MIStemperature=1.0, sampling=average@32, multi-turn=true2026.0348.38