ConceptORM / Outcome Reward Model
Outcome Reward Model
A reward model that evaluates only the correctness of the final answer. Compared with process reward models, it provides a sparser signal but requires lower training cost.
Timeline (last 90 days)
Oct 24
The outcome reward model (ORM) grants rewards only based on the correctness of the final answer; it is simple to implement but provides a sparse signal, making it hard for the model to learn correct reasoning steps from incorrect answers.
Unverified50%