List your company

Science Reasoning on Science (out-of-distribution)

65.12AccuracyCalibRL
53.898456.811759.72562.6383Feb 22, 2026

Evaluation results

MethodLinks
CalibRL2026.0265.12
LUFFY2026.0263.11
GRPO2026.0261.99
DAPO2026.0261.61
RL-PLUS2026.0261.21
SFT+GRPO2026.0254.33