Unverified50% confidenceFactExact time
在Classifier-Free Guidance机制中，模型会计算基于正面提示词和负面提示词的两个方向预测，最终生成方向在两者之间做加权插值
1
Sources
50%
Confidence
Long-term
Relevance
8/24/2026
First Seen
Sources
Related Entities
Related Claims
Unverified扩散模型在去噪过程中会将负面提示词对应的语义向量作为反向引导信号，这种机制被称为Classifier-Free Guidance的负向条件74% similarUnverifiedThe formula for Classifier-Free Guidance is: predicted noise = unconditional prediction + guidance strength × (conditional prediction − unconditional prediction), and the higher the guidance strength, the more the output deviates from the natural distribution.70% similarUnverifiedReward 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.69% similarUnverifiedFew-shot prompting leverages the in-context learning capability of the Transformer architecture, allowing the model to infer the expected output format from in-context examples alone without updating its parameters.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/793908API
curl https://kongchang.com/api/v1/knowledge/claims/793908MCP
get_claim(id=793908)