List your company

Mathematical Reasoning on GSM8K (nR2, HV_opt, HV_pes, Gap)

38.7nR2Initial
2.40411.82721.2530.673May 15, 2026

Evaluation results

MethodLinks
InitialModel=Mistral-3.2-24B, Token Budget=7.5M2026.0538.70.3520.3520
InitialModel=Qwen3-30B, Token Budget=7.5M2026.0537.40.3610.3610
NSGA-II-POModel=Qwen3-30B, Token Budget=7.5M2026.0534.30.3990.3990
MO-CAPOModel=Qwen3-30B, Token Budget=7.5M2026.05340.410.4060.4
InitialModel=GPT-OSS-120B, Token Budget=7.5M2026.0531.90.4380.4261.2
NSGA-II-POModel=Mistral-3.2-24B, Token Budget=7.5M2026.0512.20.920.8962.4
MO-CAPOModel=Mistral-3.2-24B, Token Budget=7.5M2026.05110.9430.931.3
NSGA-II-POModel=GPT-OSS-120B, Token Budget=7.5M2026.059.80.9730.925.3
MO-CAPOModel=GPT-OSS-120B, Token Budget=7.5M2026.053.81.151.1450.5