Verified65% confidenceFactExact time
Reward model training can use learning-to-rank frameworks such as the Bradley-Terry model, training a neural network to predict the probability of human preference between any two responses, taking a (prompt, response) pair as input and outputting a scalar score.
3
Sources
65%
Confidence
Long-term
Relevance
8/7/2026
First Seen
Sources
Related Claims
VerifiedDuring training, LLMs optimize parameters by maximizing the prediction probability of the next token, meaning the model learns the statistical distribution of the training corpus rather than human aesthetic judgment.74% similarUnverifiedModels can be trained with a behavioral tendency of 'when uncertain, better not to act' as a risk-avoidance mechanism encoded into weights through reward signal design73% similarVerifiedThe essence of reinforcement learning is to have the model continuously optimize to maximize a certain reward signal.73% similarVerifiedThe o-series models score each intermediate reasoning step individually using a Process Reward Model, rather than relying solely on the correctness of the final result.72% similarUnverifiedDiscriminative models learn which class a given input-output belongs to or its probability, with typical representatives including BERT and its fine-tuned variants.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/702293API
curl https://kongchang.com/api/v1/knowledge/claims/702293MCP
get_claim(id=702293)