Expired80% confidenceFactTime unknown
When training the o1 model, OpenAI employed a Process Reward Model (PRM) to perform fine-grained evaluation of intermediate reasoning steps.
7
Sources
80%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026(expired)
Sources
Related Claims
VerifiedIn 2023, OpenAI published the paper "Let's Verify Step by Step," which systematically demonstrated the superiority of process reward models (PRM) in mathematical reasoning tasks.77% similarVerifiedOpenAI's o1 series models internalize Chain of Thought as a core mechanism of model training, using reinforcement learning to teach the model to automatically unfold detailed reasoning processes.75% similarUnverifiedOpenAI argues that the training process constitutes transformative use of large volumes of data, and that the model does not simply copy original text but learns statistical patterns and linguistic structures.74% similarVerifiedIn 2020, Kaplan et al. at OpenAI revealed that model performance follows predictable power-law relationships with parameter count, training data volume, and compute budget.73% similarVerifiedA Process Reward Model (PRM) scores each intermediate step in a reasoning chain, as distinct from an Outcome Reward Model (ORM), which evaluates only the correctness of the final answer.73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/61062API
curl https://kongchang.com/api/v1/knowledge/claims/61062MCP
get_claim(id=61062)