<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Llm-Research on CuraSec</title><link>https://curasec.metacog.co.kr/tags/llm-research/</link><description>Recent content in Llm-Research on CuraSec</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Mon, 10 Aug 2026 13:39:41 +0000</lastBuildDate><atom:link href="https://curasec.metacog.co.kr/tags/llm-research/index.xml" rel="self" type="application/rss+xml"/><item><title>PDFuzzer: LLM-Driven Fuzzing Finds 31 Zero-Days in PDF Readers</title><link>https://curasec.metacog.co.kr/insights/2026-08-10-from-documentation-to-zero-day-vulnerabilities-llm-driven-fu/</link><pubDate>Mon, 10 Aug 2026 13:39:41 +0000</pubDate><guid>https://curasec.metacog.co.kr/insights/2026-08-10-from-documentation-to-zero-day-vulnerabilities-llm-driven-fu/</guid><description>&lt;ul>
&lt;li>&lt;strong>Engineer — Learn:&lt;/strong> PDFuzzer&amp;rsquo;s LLM-guided API-sequence approach found zero-days ranging from info leakage to arbitrary code execution in Adobe Acrobat, Foxit, and PDF-XChange Editor; no CVEs, patches, or exploitation signals are present yet, so watch for vendor advisories following coordinated disclosure.&lt;/li>
&lt;li>&lt;strong>SOC/IR — Learn:&lt;/strong> No active exploitation, IOCs, or TTPs to hunt for; the finding that PDF reader JavaScript engines can be exploited via chained API calls is worth noting as a future detection surface if exploitation emerges.&lt;/li>
&lt;li>&lt;strong>Leader — Skip&lt;/strong>&lt;/li>
&lt;/ul></description></item></channel></rss>