<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Inference on CyFun</title><link>https://cyfun.dev/tags/inference/</link><description>Recent content in Inference on CyFun</description><generator>Hugo -- gohugo.io</generator><language>en-us</language><lastBuildDate>Sat, 10 Oct 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://cyfun.dev/tags/inference/index.xml" rel="self" type="application/rss+xml"/><item><title>A Safer Model Made Our CTF Box Unsolvable</title><link>https://cyfun.dev/blog/safer-model-made-ctf-box-unsolvable/</link><pubDate>Sat, 10 Oct 2026 00:00:00 +0000</pubDate><guid>https://cyfun.dev/blog/safer-model-made-ctf-box-unsolvable/</guid><description>Deliberately vulnerable AI boxes, and how the choice of model, the token budget and the serving layer decide whether they can be solved.</description></item></channel></rss>