Unverified60% confidenceFactExact time
The 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.
2
Sources
60%
Confidence
Long-term
Relevance
8/6/2026
First Seen
Sources
Midjourney中世纪场景Prompt技巧拆解与实战指南
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79% similarUnverified文本引导的去噪过程本质上是Classifier-Free Guidance（CFG）机制在起作用，模型同时预测有条件和无条件的噪声估计，两者差值被放大后引导生成方向75% similarUnverified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70% similarUnverifiedThis study proposes a model-free, noise-aware Bayesian clustering framework that, from a predictive Bayesian perspective, constructs a martingale posterior over clustering structures and quantifies clustering uncertainty without explicitly specifying a model.67% similarUnverifiedClassifier-Free Guidance (CFG) is one of the most mainstream guidance techniques in diffusion models and flow matching models, discarding conditions with a certain probability during training and, during sampling, subtracting the unconditional prediction from the conditional one via a formula and amplifying the guidance strength.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/699124API
curl https://kongchang.com/api/v1/knowledge/claims/699124MCP
get_claim(id=699124)