{
  "added_at": "2026-11-02",
  "arxiv_id": "2605.29059",
  "arxiv_version_reviewed": "v1",
  "authors": [
    "Kaihua Qin",
    "Dawn Song",
    "Arthur Gervais"
  ],
  "bibtex_url": "https://awesomerea.org/papers/citations/paper-scdbench.bib",
  "code_url": null,
  "date_basis": "arXiv v1 submission date; no later version, journal reference or proceedings publication was found.",
  "date_basis_zh": "arXiv v1 首次提交日期；未找到后续版本、期刊引用信息或会议论文集出版记录。",
  "date_precision": "day",
  "description_en": "Benchmarks LLM-based EVM decompilation on 600 source-verified Ethereum Solidity contracts, sampled in easy, medium and hard thirds from 772,736 unique bytecode–source pairs after near-duplicate filtering. Each task supplies EVM assembly and optional four-byte selector hints and asks for self-contained Solidity with compiler settings. Outputs are scored in four cumulative stages: format completeness, compilation, ABI recovery, and semantic consistency, where 227,383 recorded test cases for 14,553 public functions are replayed on the recompiled output. With one compiler-feedback repair call, GPT-5.3-Codex compiled 90.3% of contracts, while the best semantic results were 29.4% of functions and 42 of 600 contracts recovered perfectly, both by Claude Opus 4.7.",
  "description_zh": "在 600 个源码已验证的以太坊 Solidity 合约上评测基于大语言模型的 EVM 反编译。这些合约在近重复过滤后，从 772,736 对唯一的字节码—源码中按简单、中等、困难各三分之一抽样。每个任务提供 EVM 汇编及可选的四字节选择器提示，要求输出自包含的 Solidity 代码与编译器设置。输出按四个累进阶段评分：格式完整性、编译、ABI 恢复，以及语义一致性；最后一阶段在重新编译的输出上重放为 14,553 个公共函数记录的 227,383 个测试用例。借助一次基于编译器反馈的修复调用，GPT-5.3-Codex 使 90.3% 的合约成功编译；语义方面的最佳结果均来自 Claude Opus 4.7，即 29.4% 的函数一致，600 个合约中有 42 个被完全恢复。",
  "doi": null,
  "entry_url": "https://awesomerea.org/papers/paper-scdbench/",
  "entry_url_zh": "https://awesomerea.org/zh/papers/paper-scdbench/",
  "evidence_urls": [
    "https://arxiv.org/abs/2605.29059v1",
    "https://arxiv.org/pdf/2605.29059v1"
  ],
  "first_arxiv_date": "2026-05-27",
  "id": "paper-scdbench",
  "json_url": "https://awesomerea.org/papers/paper-scdbench/record.json",
  "kind": "research",
  "markdown_url": "https://awesomerea.org/papers/paper-scdbench/index.md",
  "markdown_url_zh": "https://awesomerea.org/zh/papers/paper-scdbench/index.md",
  "metadata_notes": "Reviewed arXiv v1, the only version, which is labelled a preprint and has no comment, journal reference or DOI. The paper says the benchmark dataset and evaluation artifacts, including prompts, scripts and model outputs, are provided, and its broader-impact appendix describes a released dataset, but v1 gives no URL. GitHub and Hugging Face searches found no release. Compilation and replay use solc and Foundry.",
  "metadata_notes_zh": "审阅了 arXiv v1，即唯一版本，其标注为预印本，没有注释、期刊引用信息或 DOI。论文称提供基准数据集及评估产物（包括提示词、脚本和模型输出），其社会影响附录也提到已发布的数据集，但 v1 未给出网址。在 GitHub 与 Hugging Face 上检索未找到相应发布。编译与重放使用 solc 和 Foundry。",
  "pdf_url": "https://arxiv.org/pdf/2605.29059v1",
  "project_url": null,
  "publication_date": "2026-05-27",
  "publication_status": "preprint",
  "record_type": "paper",
  "runtime_tested": false,
  "scope_limitations_en": "Replay compares return data, reverts, logs and touched storage on generated inputs that reach 77.8% average code coverage; passing it is not proof of equivalence, and replay needs an Ethereum archive node. The contracts are source-verified Ethereum Solidity from compiler v0.4 to v0.8, so unverified, obfuscated, Vyper or other-chain bytecode is not covered. Three models are evaluated zero-shot through OpenRouter: Claude Opus 4.7 and GPT-5.3-Codex at high reasoning effort, and GLM-5 with and without extended thinking. Results and quoted costs describe those 2026 snapshots. Gigahorse and Heimdall-rs are scored only on selector and prototype recovery because they do not emit compilable Solidity; DiSCo and the fine-tuned decompiler of David et al. are discussed but not run.",
  "scope_limitations_zh": "重放在生成的输入上比较返回数据、回滚、日志和被访问的存储变化，这些输入平均达到 77.8% 的代码覆盖率；通过重放并不能证明等价，而且重放需要以太坊归档节点。合约均为源码已验证、编译器版本 v0.4 至 v0.8 的以太坊 Solidity 合约，因此不覆盖未验证、经混淆、Vyper 或其他链上的字节码。论文通过 OpenRouter 对三个模型进行零样本评测：高推理强度的 Claude Opus 4.7 与 GPT-5.3-Codex，以及开启和关闭扩展思考的 GLM-5。结果与所列成本仅对应这些 2026 年的模型快照。Gigahorse 与 Heimdall-rs 因不生成可编译的 Solidity，仅评测选择器与函数原型恢复；DiSCo 和 David 等人的微调反编译器仅在文中讨论，并未运行。",
  "targets": [
    "binary"
  ],
  "title": "SCDBench: A Benchmark for LLM-Based Smart Contract Decompilers",
  "title_zh": "SCDBench：基于大语言模型的智能合约反编译器基准",
  "url": "https://arxiv.org/abs/2605.29059",
  "venue": "arXiv",
  "verification": "source-reviewed",
  "verified_at": "2026-10-20",
  "why_fits_en": "Adds a shared benchmark with execution-based semantic checks for smart-contract decompilers, complementing EVM decompilation papers that evaluate on separate datasets and metrics.",
  "why_fits_zh": "为智能合约反编译器提供带有基于执行的语义检查的共同基准，与使用各自数据集和指标评测的 EVM 反编译论文互补。",
  "year": 2026
}