{
  "added_at": "2026-10-20",
  "arxiv_id": "2609.33678",
  "arxiv_version_reviewed": "v3",
  "authors": [
    "Xiaoyu Chen",
    "Lai Wei",
    "Jin Wang",
    "Xiangyu Zou",
    "Ruochen Fan",
    "Enze Luo",
    "Mingzhe Yao",
    "Jiahui Zhu",
    "Yuhua Wen",
    "Linghe Kong",
    "Weiran Huang"
  ],
  "bibtex_url": "https://awesomerea.org/papers/citations/paper-swe-game.bib",
  "code_url": null,
  "date_basis": "First arXiv submission; the reviewed version is identified separately.",
  "date_basis_zh": "采用首次 arXiv 提交日期，另列所复核版本。",
  "date_precision": "day",
  "description_en": "Evaluates construction, completion, repair and Godot-to-Unity porting against executable reference games. Evaluator-owned probes and input replays check gameplay behavior, while visual rubrics assess presentation separately.",
  "description_zh": "以可执行参考游戏评估构建、补全、修复及 Godot 到 Unity 的移植；由评测方控制的探针与输入回放检查玩法行为，视觉评分准则另行评价表现。",
  "doi": "10.48550/arxiv.2609.33678",
  "entry_url": "https://awesomerea.org/papers/paper-swe-game/",
  "entry_url_zh": "https://awesomerea.org/zh/papers/paper-swe-game/",
  "evidence_urls": [
    "https://arxiv.org/abs/2609.33678",
    "https://arxiv.org/html/2609.33678v3",
    "https://api.datacite.org/dois/10.48550/arxiv.2609.33678"
  ],
  "first_arxiv_date": "2026-09-27",
  "id": "paper-swe-game",
  "json_url": "https://awesomerea.org/papers/paper-swe-game/record.json",
  "kind": "research",
  "markdown_url": "https://awesomerea.org/papers/paper-swe-game/index.md",
  "markdown_url_zh": "https://awesomerea.org/zh/papers/paper-swe-game/index.md",
  "metadata_notes": "First submitted September 27; reviewed v3 dated October 7. No author-linked code or dataset release was verified in the reviewed manuscript.",
  "metadata_notes_zh": "首次提交为 9 月 27 日，本次复核 10 月 7 日 v3；所查手稿未核实作者关联的代码或数据发布。",
  "pdf_url": "https://arxiv.org/pdf/2609.33678v3",
  "project_url": null,
  "publication_date": "2026-09-27",
  "publication_status": "preprint",
  "record_type": "paper",
  "runtime_tested": false,
  "scope_limitations_en": "The 247 tasks derive from 41 games. Inputs vary by task; porting receives Godot source, so this is not purely black-box reconstruction. Each model uses one framework, confounding their effects. Finite scenario checks and 100-game functional validation do not establish general gameplay equivalence.",
  "scope_limitations_zh": "247 项任务来自 41 款游戏。各任务输入不同，移植任务提供 Godot 源码，因此并非纯黑盒重建。每个模型仅配一种框架，二者影响未分离；有限场景检查及 100 款游戏的功能验证不证明通用玩法等价。",
  "targets": [
    "game"
  ],
  "title": "SWE-Game: Can Coding Agents Build the Games We Want?",
  "title_zh": "SWE-Game：编程智能体能否构建符合预期的游戏？",
  "url": "https://arxiv.org/abs/2609.33678",
  "venue": "arXiv",
  "verification": "source-reviewed",
  "verified_at": "2026-10-20",
  "why_fits_en": "Adds executable checks for reference-guided game reconstruction and cross-engine behavior preservation.",
  "why_fits_zh": "补充参考引导的游戏重建及跨引擎行为保留所需的可执行检查。",
  "year": 2026
}