Unverified50% confidenceFactExact time
CFG (Classifier-Free Guidance) requires running both conditional and unconditional forward passes simultaneously at each denoising step, doubling the computation per inference step when enabled.
1
Sources
50%
Confidence
Long-term
Relevance
10/1/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedCFG performs conditional and unconditional predictions separately during inference and interpolates and amplifies them via the formula output = uncond + guidance_scale × (cond - uncond), at the cost of requiring two forward passes per step.79% similarUnverifiedCFG is a core hyperparameter in diffusion models, rooted in a 2022 research finding, and works by enhancing responsiveness to text descriptions through a weighted mixture of the predictions from the conditional and unconditional denoising paths.71% 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67% similarUnverifiedThe denoising process of diffusion models typically requires tens to hundreds of iterations (called sampling steps), each of which requires a complete forward pass through the neural network.61% similarUnverifiedDiffusionGemma采用整片并行去噪的生成机制，先铺开256个位置的噪声，然后逐轮迭代，将置信度最高的位置先固定，其余位置重新注入噪声61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/963824API
curl https://kongchang.com/api/v1/knowledge/claims/963824MCP
get_claim(id=963824)