List your company

Downstream Policy Evaluation on OLMo3 Adapt

96GSM8KSequential
93.9294.469595.54May 5, 2026

Evaluation results

MethodLinks
SequentialTraining=RC2026.059693.480.267.764.785.836.832.357.183.287.435.168.3
SequentialTraining=IQ2026.0595.992.776.566.764.68538.132548386.941.168
EVOLM2026.0595.894.586.268.567.685.337.730.554.282.386.442.269.3
GPT-4.1Prompted=true2026.0595.794.258.765.171.784.834.430.754.981.982.945.566.7
RAR2026.0595.793.877.468.662.584.83632.252.584.187.334.767.5
Scalar RMBackbone=Skywork-RM-V22026.0595.694.347.966.254.98237.325.246.256.481.628.359.7
RUBRIC-ARM2026.0595.693.378.462.364.483.835.732.856.983.187.235.967.5
Qwen3-8BPrompted=true2026.0595.594.673.758.466.883.435.331.756.984.383.945.967.5
RRD2026.0595.392.275.259.269.484.836.832.65682.688.938.567.6
RLCER2026.059494.780.567.968.684.839.230.449.669.184.337.266.7