<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Formal-Methods on AI For Human Expertise</title><link>https://aiforhumanexpertise.com/tags/formal-methods/</link><description>Recent content in Formal-Methods on AI For Human Expertise</description><generator>Hugo</generator><language>en-gb</language><lastBuildDate>Mon, 06 Jul 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://aiforhumanexpertise.com/tags/formal-methods/index.xml" rel="self" type="application/rss+xml"/><item><title>Constrained Generation Is Symbolic AI Smuggled Into LLMs</title><link>https://aiforhumanexpertise.com/blog/constrained-generation-is-symbolic-ai/</link><pubDate>Mon, 06 Jul 2026 00:00:00 +0000</pubDate><guid>https://aiforhumanexpertise.com/blog/constrained-generation-is-symbolic-ai/</guid><description>&lt;h2 id="the-toolkit-most-practitioners-already-reach-for"&gt;The Toolkit Most Practitioners Already Reach For&lt;/h2&gt;
&lt;p&gt;Anyone who has shipped a language-model application beyond the prototype stage has, at some point, used constrained generation. The shape of the encounter is consistent. The model produces text that is almost right and occasionally wrong in ways the consumer cannot tolerate. Someone on the team reaches for one of a small set of tools: Outlines, lm-format-enforcer, the JSON-schema mode in their model API, a regex post-filter, a structured-output decoder. The wrong answers stop. The system ships.&lt;/p&gt;</description></item></channel></rss>