From 1b33cd7e9fc5f9d2676b377eaf0cfb39b95e533e Mon Sep 17 00:00:00 2001 From: Yan Lin Date: Fri, 13 Feb 2026 17:24:22 +0100 Subject: [PATCH] publish the new post --- content/ml-tech/diffusion-guidance/index.md | 1 - 1 file changed, 1 deletion(-) diff --git a/content/ml-tech/diffusion-guidance/index.md b/content/ml-tech/diffusion-guidance/index.md index 49e25c3..82b3e39 100644 --- a/content/ml-tech/diffusion-guidance/index.md +++ b/content/ml-tech/diffusion-guidance/index.md @@ -2,7 +2,6 @@ title = "From Soft Guidance to Hard Constraints on Diffusion Sampling" date = 2026-02-13 description = "" -draft = true +++ Diffusion models, or more broadly speaking, both score-matching and flow-matching models, are the foundational frameworks for building generative models these days. Compared to previous frameworks including VAEs and GANs, they are actually capable of generating high-quality data, ~like NSFW images that you might or might not want to jerk off to~.