{
  "added_at": "2026-10-16",
  "arxiv_id": "2609.22308",
  "arxiv_version_reviewed": "v1",
  "authors": [
    "Boyu Qiao",
    "Zixin Tang",
    "Xiaoshuai Hao",
    "Wenbo Li"
  ],
  "bibtex_url": "https://awesomerea.org/papers/citations/paper-gamereplica.bib",
  "code_url": null,
  "date_basis": "arXiv v1 submission history",
  "date_basis_zh": "arXiv v1 的首次提交日期。",
  "date_precision": "day",
  "description_en": "Evaluates whether vision-language agents can infer hidden game rules through screenshots and active interaction, then build runnable single-file HTML replicas across 125 tasks from 25 web games, each with five difficulty levels.",
  "description_zh": "通过来自 25 款网页游戏、各设五个难度级别的 125 个任务，评估视觉语言智能体能否通过截图和主动交互推断隐藏规则，并构建可运行的单文件 HTML 游戏复刻。",
  "doi": "10.48550/arXiv.2609.22308",
  "entry_url": "https://awesomerea.org/papers/paper-gamereplica/",
  "entry_url_zh": "https://awesomerea.org/zh/papers/paper-gamereplica/",
  "evidence_urls": [
    "https://arxiv.org/abs/2609.22308",
    "https://arxiv.org/html/2609.22308v1",
    "https://qqqqqqby.github.io/GameReplica/"
  ],
  "first_arxiv_date": "2026-09-15",
  "id": "paper-gamereplica",
  "json_url": "https://awesomerea.org/papers/paper-gamereplica/record.json",
  "kind": "research",
  "markdown_url": "https://awesomerea.org/papers/paper-gamereplica/index.md",
  "markdown_url_zh": "https://awesomerea.org/zh/papers/paper-gamereplica/index.md",
  "metadata_notes": "The official project page marks the work as under review. arXiv reports the v1 submission on 2026-09-15 and displays its arXiv DOI as pending registration. The project link is a showcase with demos, not a verified release of the full evaluation framework.",
  "metadata_notes_zh": "官方项目页将该工作标注为审稿中。arXiv 记载 v1 于 2026-09-15 提交，并将其 arXiv DOI 标注为待注册。项目链接为含演示的展示页，不代表已核实发布完整评测框架。",
  "pdf_url": "https://arxiv.org/pdf/2609.22308v1",
  "project_url": "https://qqqqqqby.github.io/GameReplica/",
  "publication_date": "2026-09-15",
  "publication_status": "preprint",
  "record_type": "paper",
  "runtime_tested": false,
  "scope_limitations_en": "The tasks cover rule-driven web puzzles and board games, not native or commercial game reconstruction. Rule, implementation, and visual scores use LLM judges and do not prove behavioral equivalence. The official site provides demos; a benchmark-code repository was not verified.",
  "scope_limitations_zh": "任务覆盖以规则为核心的网页解谜和棋盘游戏，不代表原生或商业游戏重建能力。规则、实现和视觉评分使用 LLM 评审，不能证明行为等价。官方网站提供演示，尚未核实基准代码仓库。",
  "targets": [
    "game",
    "website"
  ],
  "title": "GameReplica: A Benchmark for Black-Box Visual Game Replication by Vision-Language Agents",
  "title_zh": "GameReplica：视觉语言智能体的黑盒视觉游戏复刻基准",
  "url": "https://arxiv.org/abs/2609.22308",
  "venue": "arXiv",
  "verification": "source-reviewed",
  "verified_at": "2026-10-16",
  "why_fits_en": "Directly evaluates active black-box specification recovery and executable game reconstruction, filling the bibliography’s game-replication gap.",
  "why_fits_zh": "直接评估主动黑盒规格恢复与可执行游戏重建，补充书目中的游戏复刻研究。",
  "year": 2026
}