一句话：提出 AST-Align 对齐算法并构建 R³-Bench 数据集，使结构体字段真值提升四倍，为二进制符号恢复提供评估基准。
逆向工程中的符号恢复旨在从编译后二进制中恢复变量与数据结构信息。学习式方法依赖二进制表达式与源码精确对齐的数据集，但现有对齐技术难以处理复杂数据访问模式，导致训练数据不完整并限制模型性能。本文提出 AST-Align，跨 x86/ARM 架构与 C/C++/Rust 语言统一变量及结构体访问表达式对齐，生成的结构体字段真值达先前方法四倍。基于该算法构建 R³-Bench：元数据丰富、可扩展，具有明确项目纳入标准与可复现流程，覆盖多架构超千万函数。评估从 n-gram 到 LLM 建立基线，通用 LLM 初始表现差，但在合适示例下显著提升。
Symbol recovery in reverse engineering is crucial for restoring variable and data structure information in compiled binaries. While learning-based methods have shown promise in recovering both semantic information (names and types) and syntactic information (shapes), they require comprehensive datasets where expressions in binary code are precisely aligned with their source code equivalents. Current techniques for generating such alignments struggle with complex data access patterns, resulting in incomplete training data and consequently hampering model performance and recovery accuracy. We present AST-Align, a novel technique unifying alignment of variables and struct access expressions across multiple architectures (x86 and ARM) and languages (C/C++/Rust). AST-Align significantly improves the number of generated ground truths, capturing four times more struct fields than previous methods. Using this algorithm, we develop R3-Bench, a metadata-rich, extensible dataset with explicit project inclusion criteria and reproducible processing pipeline, comprising over 10 million functions across multiple architectures. Our evaluation establishes baseline performance by testing various approaches from n-gram models to Large Language Models. The results show that while general LLMs initially perform poorly, their effectiveness dramatically improves with proper demonstration. R3-Bench provides a robust foundation for assessing model capabilities and serves as a valuable reference for future symbol recovery research.
DOI 原文 ·
@inproceedings{paperbot9334,
title = {R
<sup>3</sup>
-Bench: Reproducible Real-world Reverse Engineering Dataset for Symbol Recovery},
author = {Muzhi Yu and Zhengran Zeng and Wei Ye and Jinan Sun and Xiaolong Bai and Shikun Zhang},
booktitle = {ASE},
year = {2025},
doi = {10.1109/ase63991.2025.00026}
}