OVM, Outcome-supervised Value Models for Planning in Mathematical Reasoning

OVM (Outcome-supervised Value Models) is a research model for planning in mathematical reasoning tasks. It uses outcome supervision to train value functions that evaluate the correctness of intermediate reasoning steps, enabling more effective search and planning during problem solving. The model is described in the paper 'OVM, Outcome-supervised Value Models for Planning in Mathematical Reasoning' and is available as open-source software.