Unverified50% confidenceFactExact time
负样本提示最早在 Stable Diffusion 社区中被系统性开发，其技术实现依赖分类器自由引导（Classifier-Free Guidance, CFG）机制
1
Sources
50%
Confidence
Long-term
Relevance
7/6/2026
First Seen
Sources
Taste-Skill: The Open-Source AI Taste Engine That Fights Generic AI-Generated Content
githubGitHub Trending7/5/2026
Related Claims
UnverifiedIn open-source models such as Stable Diffusion, the negative prompt is a separate input field used to tell the model to actively avoid certain visual features.70% similarUnverifiedStable Diffusion的核心代码最初由一个小团队完成，LLaMA的开源推动了社区推理优化浪潮66% similarUnverifiedControlNet connects a pretrained Stable Diffusion model with a conditioning branch through zero convolution layers, allowing precise control over generated results via additional conditional inputs.61% similarUnverifiedStable Diffusion supports using weight syntax (such as `(keyword:1.5)`) to precisely control the influence strength of a specific element, combined with Negative Prompts to exclude unwanted elements.60% similarUnverifiedClassifier-free guidance (CFG) was proposed by Jonathan Ho and Tim Salimans in 2022; by randomly dropping text conditioning during training, the model learns both conditional and unconditional generation.60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/115649API
curl https://kongchang.com/api/v1/knowledge/claims/115649MCP
get_claim(id=115649)