Glossary · term
Outcome Reward Model (ORM)
A reward model that evaluates only the final result of a response, not the individual reasoning steps. It learns from "correct/incorrect solution" pairs, so it is cheaper and simpler than a Process Reward Model (PRM), but it is worse at distinguishing correct reasoning from getting the answer right by chance. It is often a counterpoint to PRM.
Training2024–2025Wave 2 · 2024Maturity: 5/5
Maturity rationale
GPAI Code of Practice — within the EU AI Act framework
Cite or link to this resource · CC BY 4.0
Free to quote, link, and reuse with attributionReferencing this analysis in documentation, research, or an engineering write-up? Copy a formatted link or citation below:
Show citation text & canonical URL
Markdown:
[Outcome Reward Model (ORM) — AI Glossary — Skills Intelligence](https://www.skillsintelligence.tools/glossary/term/outcome-reward-model-orm)APA:
Skills Intelligence. (2024). Outcome Reward Model (ORM) — AI Glossary. Skills Intelligence AI Glossary. Retrieved from https://www.skillsintelligence.tools/glossary/term/outcome-reward-model-orm