OpenAI 推出了 sCM，一种新的连续时间一致性模型方法，显著加快了生成式 AI 的采样速度。该方法简化和稳定了训练，使模型能够仅用两步生成与扩散模型相当的高质量样本，速度提升了约 50 倍。研究表明，sCMs 可以有效地与教师扩散模型进行扩展，并在 ImageNet 等基准测试中以更低的计算成本取得最先进的成果。 AI
排序理由 OpenAI 发布了关于一类名为一致性模型的新型生成模型的研究，详细介绍了改进和扩展技术。
AI 生成摘要 · Google Gemini · 来自 4 个来源。 我们如何撰写摘要 →
We’ve simplified, stabilized, and scaled continuous-time consistency models, achieving comparable sample quality to leading diffusion models, while using only two sampling steps.
Consistency models are a nascent family of generative models that can sample high quality data in one step without the need for adversarial training.
Diffusion models have significantly advanced the fields of image, audio, and video generation, but they depend on an iterative sampling process that causes slow generation.
**Model distillation** significantly accelerates diffusion models, enabling near real-time image generation with only 1-4 sampling steps, as seen in **BlinkShot** and **Flux Schnell**. Research led by **Yang Song** introduced **simplified continuous-time consistency models (sCMs)…