arXiv cs.AI· Ritika Poddar, Aidan Fennelly, Deirdre Murphy, Eric Wolos, Corre L. Steele, Michael Doyle·· 3 小时前AI 评分45
BUSTA：教会小模型推理二进制执行
BUSTA: Teaching Small Language Models to Reason About Binary Execution
AI 导读
arXiv 论文 BUSTA 提出用反向基本块执行任务训练小语言模型掌握逆向工程推理能力：给定 x86-64 基本块和目标输出寄存器状态，模型预测输入寄存器值。经 GRPO 微调后，Qwen3-1.7B 的寄存器预测准确率从 73.2% 升至 85.3%，Nemotron-3-Nano 从 63.3% 升至 95.2%，逼近前沿模型最高 99.2% 的成绩。微调后的两个模型及训练数据集均已开源。
正文 · 原文
Abstract:Large Language Models (LLM) are increasingly applied to Reverse Engineering (RE) workflows. RE, however, differs from common LLM coding tasks as it requires models to reason about low-level program behavior. Furthermore, the deployment of tooling in resource-constrained sandboxed RE environments limits the ability to use frontier-scale solutions. We examine whether a targeted LLM training approach can transfer distinct RE skills to open models via inverse basic-block execution: given an x86-64 basic block and a desired output register state, the model predicts the input register values that produce it. Fine-tuning \texttt{Qwen3-1.7B} and \texttt{Nemotron-3-Nano} with Group Relative Policy Optimization (GRPO) raises knowable-register prediction accuracy from 73.2\% to 85.3\% and from 63.3\% to 95.2\%, respectively. Frontier models still score highest (up to 99.2\%), but our fine-tuned \texttt{Nemotron-3-Nano} comes within four points at a fraction of the parameter count. We release both fine-tuned models and the training dataset.
| Subjects: | Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Programming Languages (cs.PL) |
| Cite as: | arXiv:2610.36288 [cs.CR] |
| (or arXiv:2610.36288v1 [cs.CR] for this version) | |
| https://doi.org/10.48550/arXiv.2610.36288 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Michael Doyle [view email]
[v1]
Mon, 26 Oct 2026 21:11:48 UTC (23 KB)
来源：arXiv cs.AI · arxiv.org