{
  "added_at": "2026-10-17",
  "arxiv_id": "2505.12668",
  "arxiv_version_reviewed": "v2",
  "authors": [
    "Hanzhuo Tan",
    "Xiaolong Tian",
    "Hanrui Qi",
    "Jiaming Liu",
    "Zuchen Gao",
    "Siyi Wang",
    "Qi Luo",
    "Jing Li",
    "Yuqun Zhang"
  ],
  "bibtex_url": "https://awesomerea.org/papers/citations/paper-decompile-bench.bib",
  "code_url": "https://github.com/albertan017/LLM4Decompile",
  "date_basis": "Publication year in the official NeurIPS proceedings and publisher-deposited Crossref metadata; no finer publication date is asserted.",
  "date_basis_zh": "采用 NeurIPS 官方论文集及出版商在 Crossref 登记的出版年份，不推定更精确的出版日期。",
  "date_precision": "year",
  "description_en": "Introduces two million C/C++ binary-source function pairs using a Compile-Trace-Filter pipeline, with an evaluation suite spanning translated HumanEval/MBPP tasks and recent GitHub functions.",
  "description_zh": "通过 Compile-Trace-Filter 流程构建 200 万组 C/C++ 二进制与源码函数对，并提供涵盖转写为 C/C++ 的 HumanEval、MBPP 任务及近期 GitHub 函数的评测集。",
  "doi": "10.52202/085713-0177",
  "entry_url": "https://awesomerea.org/papers/paper-decompile-bench/",
  "entry_url_zh": "https://awesomerea.org/zh/papers/paper-decompile-bench/",
  "evidence_urls": [
    "https://papers.nips.cc/paper_files/paper/2025/hash/079cf13ae174c31f148207d94d213bdc-Abstract-Datasets_and_Benchmarks_Track.html",
    "https://papers.nips.cc/paper_files/paper/2025/file/079cf13ae174c31f148207d94d213bdc-Paper-Datasets_and_Benchmarks_Track.pdf",
    "https://doi.org/10.52202/085713-0177",
    "https://api.crossref.org/works/10.52202/085713-0177",
    "https://arxiv.org/abs/2505.12668",
    "https://arxiv.org/html/2505.12668v2",
    "https://github.com/albertan017/LLM4Decompile/blob/85b364bf093eb2ee4f3687cfe38a203fca89f23e/decompile-bench/readme.md",
    "https://github.com/albertan017/LLM4Decompile/blob/85b364bf093eb2ee4f3687cfe38a203fca89f23e/LICENSE",
    "https://github.com/albertan017/LLM4Decompile/blob/85b364bf093eb2ee4f3687cfe38a203fca89f23e/LICENSE-MODEL",
    "https://huggingface.co/datasets/LLM4Binary/decompile-bench/blob/4b708c2211cd7d4af403675db56322aa4ed7050c/README.md",
    "https://huggingface.co/datasets/LLM4Binary/decompile-bench/blob/4b708c2211cd7d4af403675db56322aa4ed7050c/LICENSE"
  ],
  "first_arxiv_date": "2025-05-19",
  "id": "paper-decompile-bench",
  "json_url": "https://awesomerea.org/papers/paper-decompile-bench/record.json",
  "kind": "research",
  "markdown_url": "https://awesomerea.org/papers/paper-decompile-bench/index.md",
  "markdown_url_zh": "https://awesomerea.org/zh/papers/paper-decompile-bench/index.md",
  "metadata_notes": "Author order follows the final PDF, which puts Zuchen Gao before Siyi Wang; the proceedings landing page and Crossref reverse them. arXiv v2 was revised on 2025-10-19. The artifact is the decompile-bench subdirectory of the existing LLM4Decompile repository. Code has an MIT license; the dataset declares CC0-1.0, while model weights have separate terms. These declarations do not replace source-project license review.",
  "metadata_notes_zh": "作者顺序以正式 PDF 为准，其中 Zuchen Gao 位于 Siyi Wang 之前；论文集网页与 Crossref 的顺序相反。arXiv v2 于 2025-10-19 更新。产物位于已有 LLM4Decompile 仓库的 decompile-bench 子目录。代码采用 MIT 许可，数据集声明 CC0-1.0，模型权重另有条款；这些声明不能替代对源项目许可的核查。",
  "pdf_url": "https://papers.nips.cc/paper_files/paper/2025/file/079cf13ae174c31f148207d94d213bdc-Paper-Datasets_and_Benchmarks_Track.pdf",
  "project_url": null,
  "publication_date": null,
  "publication_status": "conference-paper",
  "record_type": "paper",
  "runtime_tested": false,
  "scope_limitations_en": "This is function-level decompilation. Re-execution tests cover HumanEval/MBPP; GitHub2025 provides no unit tests. Inlining can prevent one-to-one source correspondence, and the fine-tuning study uses 10% of the corpus. These results do not establish whole-program recovery.",
  "scope_limitations_zh": "该工作研究函数级反编译。重新执行测试覆盖 HumanEval、MBPP；GitHub2025 未提供单元测试。内联可能使二进制与源码无法一一对应，微调实验仅使用 10% 的语料。这些结果不能证明整程序恢复能力。",
  "targets": [
    "binary"
  ],
  "title": "Decompile-Bench: Million-Scale Binary-Source Function Pairs for Real-World Binary Decompilation",
  "title_zh": "Decompile-Bench：面向真实二进制反编译的百万级二进制与源码函数对",
  "url": "https://papers.nips.cc/paper_files/paper/2025/hash/079cf13ae174c31f148207d94d213bdc-Abstract-Datasets_and_Benchmarks_Track.html",
  "venue": "NeurIPS 2025 Datasets and Benchmarks",
  "verification": "source-reviewed",
  "verified_at": "2026-10-16",
  "why_fits_en": "Adds a training corpus and evaluation artifact for LLM decompilation, distinct from Gao et al.’s existing DecompileBench entry.",
  "why_fits_zh": "补充 LLM 反编译的训练语料与评测产物，不同于现有 Gao 等人的 DecompileBench 条目。",
  "year": 2025
}