{
  "added_at": "2026-10-16",
  "arxiv_id": "2403.05286",
  "arxiv_version_reviewed": "v3",
  "authors": [
    "Hanzhuo Tan",
    "Qi Luo",
    "Jing Li",
    "Yuqun Zhang"
  ],
  "bibtex_url": "https://awesomerea.org/papers/citations/paper-llm4decompile.bib",
  "code_url": "https://github.com/albertan017/LLM4Decompile",
  "date_basis": "ACL proceedings publication month",
  "date_basis_zh": "ACL 会议论文集的出版月份。",
  "date_precision": "month",
  "description_en": "Trains language models for assembly-to-C decompilation and for refining Ghidra pseudocode, evaluating whether recovered functions can be recompiled and executed correctly.",
  "description_zh": "训练语言模型将汇编反编译为 C，并改进 Ghidra 伪代码，以恢复函数能否重新编译并正确执行为评估目标。",
  "doi": "10.18653/v1/2024.emnlp-main.203",
  "entry_url": "https://awesomerea.org/papers/paper-llm4decompile/",
  "entry_url_zh": "https://awesomerea.org/zh/papers/paper-llm4decompile/",
  "evidence_urls": [
    "https://aclanthology.org/2024.emnlp-main.203/",
    "https://aclanthology.org/2024.emnlp-main.203.pdf",
    "https://arxiv.org/abs/2403.05286",
    "https://github.com/albertan017/LLM4Decompile",
    "https://github.com/albertan017/LLM4Decompile/blob/85b364bf093eb2ee4f3687cfe38a203fca89f23e/README.md"
  ],
  "first_arxiv_date": "2024-03-08",
  "id": "paper-llm4decompile",
  "json_url": "https://awesomerea.org/papers/paper-llm4decompile/record.json",
  "kind": "research",
  "markdown_url": "https://awesomerea.org/papers/paper-llm4decompile/index.md",
  "markdown_url_zh": "https://awesomerea.org/zh/papers/paper-llm4decompile/index.md",
  "metadata_notes": "Publication date is the conference month; the first arXiv submission date is listed separately.",
  "metadata_notes_zh": "出版日期采用会议论文集月份；arXiv 首次提交日期单独记录。",
  "pdf_url": "https://aclanthology.org/2024.emnlp-main.203.pdf",
  "project_url": null,
  "publication_date": "2024-11",
  "publication_status": "conference-paper",
  "record_type": "paper",
  "runtime_tested": false,
  "scope_limitations_en": "The paper limits its experiments to C on x86 and function-level benchmarks. Re-executability on sampled tests does not establish whole-program equivalence or recovery of the original source. Later repository releases are separate from the 2024 paper.",
  "scope_limitations_zh": "论文实验限定于 x86 平台的 C 代码和函数级基准。通过抽样测试重新执行，不代表整程序等价或恢复了原始源码。仓库后续版本应与 2024 年论文区分。",
  "targets": [
    "binary"
  ],
  "title": "LLM4Decompile: Decompiling Binary Code with Large Language Models",
  "title_zh": "LLM4Decompile：使用大语言模型反编译二进制代码",
  "url": "https://aclanthology.org/2024.emnlp-main.203/",
  "venue": "EMNLP 2024",
  "verification": "source-reviewed",
  "verified_at": "2026-10-16",
  "why_fits_en": "Direct evidence for LLM decompilation and a concrete baseline for evaluating agent-generated or refined pseudocode.",
  "why_fits_zh": "直接研究 LLM 反编译，可作为评估智能体生成或改进伪代码的具体基线。",
  "year": 2024
}