← 研踪 PaperTrailarXiv:2605.09961 · v2
Towards LLM-Based Analysis of Virtualization-Obfuscated Code through Automated Data Generation
首次提交 2026-05-11 12:15
版本更新 2026-11-09 23:53
雷达观测 2026-11-10 15:22
摘要
Virtualization-based obfuscation produces extremely large and structurally complex binaries, posing challenges for LLM-based analysis due to input size limits and the need for large-scale labeled data. We address this by focusing on structural rather than full semantic analysis. Obfuscated binaries are decomposed into the largest semantically coherent units that fit within LLM constraints and are labeled according to their structural roles. We implement a static analysis framework to automate labeling and enable large-scale dataset generation. Our prototype shows strong performance on real-world virtualization obfuscators.
作者备注：arXiv admin note: This submission has been withdrawn because it does not meet arXiv's research content quality standards
观测到的版本
相邻记录可能跳过中间版本；首次发现 v2 及以上也会标为版本更新。
v2 · 2026-11-09 23:53 · 最新观测版本
Towards LLM-Based Analysis of Virtualization-Obfuscated Code through Automated Data Generation
Virtualization-based obfuscation produces extremely large and structurally complex binaries, posing challenges for LLM-based analysis due to input size limits and the need for large-scale labeled data. We address this by focusing on structural rather than full semantic analysis. Obfuscated binaries are decomposed into the largest semantically coherent units that fit within LLM constraints and are labeled according to their structural roles. We implement a static analysis framework to automate labeling and enable large-scale dataset generation. Our prototype shows strong performance on real-world virtualization obfuscators.
查看此版本解读 →查看此版本原文 ↗