{
  "added_at": "2026-10-16",
  "arxiv_id": "2108.06363",
  "arxiv_version_reviewed": "v1",
  "authors": [
    "Qibin Chen",
    "Jeremy Lacomis",
    "Edward J. Schwartz",
    "Claire Le Goues",
    "Graham Neubig",
    "Bogdan Vasilescu"
  ],
  "bibtex_url": "https://awesomerea.org/papers/citations/paper-dirty.bib",
  "code_url": "https://github.com/CMUSTRUDEL/DIRTY",
  "date_basis": "USENIX proceedings BibTeX publication month",
  "date_basis_zh": "USENIX 论文集 BibTeX 记录中的出版月份。",
  "date_precision": "month",
  "description_en": "Introduces DIRTY, a Transformer model that recommends variable names and types from decompiled code and layout information, together with the DIRT corpus for training and evaluation.",
  "description_zh": "提出 Transformer 模型 DIRTY，根据反编译代码和布局信息推荐变量名与类型，并提供用于训练和评估的 DIRT 语料。",
  "doi": null,
  "entry_url": "https://awesomerea.org/papers/paper-dirty/",
  "entry_url_zh": "https://awesomerea.org/zh/papers/paper-dirty/",
  "evidence_urls": [
    "https://www.usenix.org/conference/usenixsecurity22/presentation/chen-qibin",
    "https://arxiv.org/abs/2108.06363",
    "https://github.com/CMUSTRUDEL/DIRTY",
    "https://github.com/CMUSTRUDEL/DIRTY/blob/5f8e9494d52ef8788f0834342097f36462eeb2eb/README.md",
    "https://edmcman.github.io/papers/usenix22.pdf"
  ],
  "first_arxiv_date": "2021-08-13",
  "id": "paper-dirty",
  "json_url": "https://awesomerea.org/papers/paper-dirty/record.json",
  "kind": "research",
  "markdown_url": "https://awesomerea.org/papers/paper-dirty/index.md",
  "markdown_url_zh": "https://awesomerea.org/zh/papers/paper-dirty/index.md",
  "metadata_notes": "Year and publication month come from the final USENIX conference record; the arXiv date is not substituted for publication.",
  "metadata_notes_zh": "年份与出版月份来自正式 USENIX 会议记录，不以较早的 arXiv 提交日期代替。",
  "pdf_url": "https://arxiv.org/pdf/2108.06363",
  "project_url": "https://dirtdirty.github.io/explorer.html",
  "publication_date": "2022-08",
  "publication_status": "conference-paper",
  "record_type": "paper",
  "runtime_tested": false,
  "scope_limitations_en": "The original pipeline is built around C programs and Hex-Rays output. Predicted names and types are learned hypotheses, not guaranteed original symbols. The published paper is from 2022, although its arXiv preprint first appeared in 2021.",
  "scope_limitations_zh": "原始流程围绕 C 程序和 Hex-Rays 输出构建。预测名称与类型是模型推断，不保证与原始符号一致。正式论文发表于 2022 年，arXiv 预印本首次提交于 2021 年。",
  "targets": [
    "binary"
  ],
  "title": "Augmenting Decompiler Output with Learned Variable Names and Types",
  "title_zh": "利用学习得到的变量名称与类型增强反编译输出",
  "url": "https://www.usenix.org/conference/usenixsecurity22/presentation/chen-qibin",
  "venue": "USENIX Security 2022",
  "verification": "source-reviewed",
  "verified_at": "2026-10-16",
  "why_fits_en": "Extends name recovery to types, making it directly relevant to assessing semantic annotations added by binary-analysis assistants.",
  "why_fits_zh": "将名称恢复扩展到类型恢复，直接帮助评估二进制分析助手添加的语义注释。",
  "year": 2022
}