<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Vulnerability Repair | Hulin Wang</title><link>https://www.hulinwang.me/tag/vulnerability-repair/</link><atom:link href="https://www.hulinwang.me/tag/vulnerability-repair/index.xml" rel="self" type="application/rss+xml"/><description>Vulnerability Repair</description><generator>Wowchemy (https://wowchemy.com)</generator><language>en-us</language><lastBuildDate>Wed, 01 Apr 2026 00:00:00 +0000</lastBuildDate><image><url>https://www.hulinwang.me/media/sharing.png</url><title>Vulnerability Repair</title><link>https://www.hulinwang.me/tag/vulnerability-repair/</link></image><item><title>Autonomous Vulnerability Patching Agent — DARPA AIxCC</title><link>https://www.hulinwang.me/project/aixcc-patching-agent/</link><pubDate>Fri, 08 Aug 2025 00:00:00 +0000</pubDate><guid>https://www.hulinwang.me/project/aixcc-patching-agent/</guid><description>&lt;p>&lt;strong>Role:&lt;/strong> Patching Component Lead, Team Shellphish (Aug 2023 – Aug 2025)&lt;/p>
&lt;p>The DARPA Artificial Intelligence Cyber Challenge (AIxCC) is a global competition for automated vulnerability discovery and repair in open-source and critical-infrastructure software. Team Shellphish was named one of 7 finalists in Aug 2024 (&lt;strong>$2M&lt;/strong> finalist award) and placed &lt;strong>5th&lt;/strong> in the final round in Aug 2025.&lt;/p>
&lt;h3 id="vulnerability-patching-agent">Vulnerability patching agent&lt;/h3>
&lt;ul>
&lt;li>Led development of a multi-agent vulnerability repair system integrating autonomous triage, patch generation, and validation across distributed services.&lt;/li>
&lt;li>Built a feedback loop combining fuzzing, proof-of-vulnerability re-tests, and deterministic regression tests, so that only patches that eliminate the vulnerability &lt;em>and&lt;/em> preserve functionality are accepted — &lt;strong>&amp;gt;95% patch correctness&lt;/strong> in the final scored rounds.&lt;/li>
&lt;li>Triaged and patched &lt;strong>30+ C/Java vulnerabilities&lt;/strong> in projects including curl, NGINX, and Apache ZooKeeper, covering buffer overflows, use-after-free, command injection, and path traversal.&lt;/li>
&lt;/ul>
&lt;h3 id="proof-of-concept-generation-agent">Proof-of-concept generation agent&lt;/h3>
&lt;ul>
&lt;li>Developed a multi-agent system that combines static-analysis results (CodeQL) and code context to generate inputs targeting specific vulnerabilities.&lt;/li>
&lt;li>Built a reachability-based program-slicing pipeline to focus agents on relevant code paths, using coverage-guided search to bypass blocking logic.&lt;/li>
&lt;li>Generated PoCs in &lt;strong>~5 minutes&lt;/strong> on average, triggering 10 vulnerabilities in ZooKeeper, Tika, and jsoup — &lt;strong>five of which Jazzer missed after 8 hours&lt;/strong> of fuzzing.&lt;/li>
&lt;/ul></description></item><item><title>Root-Cause-Driven Vulnerability Repair</title><link>https://www.hulinwang.me/project/root-cause-repair/</link><pubDate>Wed, 01 Apr 2026 00:00:00 +0000</pubDate><guid>https://www.hulinwang.me/project/root-cause-repair/</guid><description>&lt;p>&lt;strong>Status:&lt;/strong> First-author paper, under submission. Ph.D. research at ASU SEFCOM.&lt;/p>
&lt;ul>
&lt;li>Designed a two-stage repair pipeline combining static and dynamic analysis to locate potential root causes, capturing defects that span multiple code locations and indirect calls missed by static-only tools such as CodeQL.&lt;/li>
&lt;li>Supplied root-cause candidates to a lightweight patching agent with repository-search and file-editing tools, so repairs are informed by the relevant code context.&lt;/li>
&lt;li>Built a manually annotated patch-quality dataset covering &lt;strong>178 C/C++ vulnerabilities&lt;/strong> to compare repairs generated by multiple agents.&lt;/li>
&lt;li>Improved patch quality by &lt;strong>20%&lt;/strong> over baseline methods, validated by majority vote of three human experts.&lt;/li>
&lt;/ul>
&lt;h3 id="related-llm-assisted-fuzzing-for-occluded-bugs">Related: LLM-assisted fuzzing for occluded bugs&lt;/h3>
&lt;p>An LLM-assisted patching + AFL++ pipeline that surfaces vulnerabilities masked by more easily triggered crashes. Automated crash clustering, root-cause hints, LLM-guided edits, and re-fuzzing reduced false positives from 90% to 30% and found &lt;strong>20+ distinct occluded vulnerabilities per target&lt;/strong> in Objdump, FFmpeg, and GPAC.&lt;/p></description></item></channel></rss>