List your company

Tutoring Response Generation on MathDial

0.892RewardBayesian Decision Rule
0.805680.828090.85050.87291Jun 29, 2026

Evaluation results

MethodLinks
Bayesian Decision RuleGenerator=TutorRL-7B, Conformal Pruning=false2026.060.89254.375.4
Bayesian Decision RuleGenerator=TutorRL-7B, Conformal Pruning=true2026.060.89254.375.2
SamplingGenerator=TutorRL-7B2026.060.858—71.6
Minmax Decision RuleGenerator=TutorRL-7B2026.060.80939.362.2