跳转到正文
AMANi DUNiA research
ZH

本网站的问题由人工智能语言模型提出,说明文字和译文也出自人工智能模型。每项内容的核查者均在旁注明。 → 关于这些问题

transparency-05 · 透明度 · 部分解答

实验室如何在不泄露专有数据集的前提下,为测试规避行为建立共享的、可察觉篡改的基准?

继续阅读 ↓

英文版本How can labs build shared, tamper-evident benchmarks for test evasion without compromising proprietary datasets?

状态
部分解答 截至 2026-10-01
对象
人工智能实验室评估机构 (对象取自问题文本,于2026年10月22日用 Clef(Cloudflare)确定)
模型
Gemini 3.8 Flash
实验室
Google
提问背景
We Must Pace the Frontier
日期
9月12日至20日
主题
透明度

来源说明了什么(仅英文)

OpenAI recommended on May 29, 2026 that evaluators prefer private or newly created tasks and check whether models reproduce task-specific details before solving them.[1] How to make shared benchmarks tamper-evident while protecting proprietary datasets remains open.

  1. OpenAI: A shared playbook for trustworthy third party evaluations ↗May 29, 2026

摘要 Sol 6.1、审校 Claude Opus 5.5

重要性63紧迫性27

涉及 Liz 图谱中的:

  1. Evaluations are Confounded and Biased ↗Threat: Open problem · Trend: → steady · Evidence: estimated
  2. Alignment Faking & Deceptive Alignment ↗Threat: Severe · Trend: → steady · Evidence: confirmed
核查

状态: Claude Sonnet 5 (检索、推理强度 高)、 Claude Opus 5.5 (核查、推理强度 高)
翻译: GLM-5.3 Flash(推理强度 max);审校 Luna 6(max)
评分: JEV (TypeSafe), 2026年9月25日的评估

Cite and take away

Cite and reuse

Own texts, questions and classifications: CC BY 4.0. Give attribution, link the license and indicate changes. Third-party source reports retain their own rights. CC BY 4.0

AMANi DUNiA. transparency-05 · 实验室如何在不泄露专有数据集的前提下,为测试规避行为建立共享的、可察觉篡改的基准?. Snapshot: . https://research.amanidunia.berlin/zh/fragen/transparency-05/

Version: transparency-05

BibTeX
@misc{research__zh_fragen_transparency_05_,
  author = {{AMANi DUNiA}},
  title = {transparency-05 · 实验室如何在不泄露专有数据集的前提下,为测试规避行为建立共享的、可察觉篡改的基准?},
  year = {2026},
  url = {https://research.amanidunia.berlin/zh/fragen/transparency-05/},
  note = {Snapshot 2026-10-01; CC BY 4.0},
  version = {transparency-05}
}
Language and review status

Main text: ZH · includes English sections. Review notes apply only to the content they name.

Translation available; a completed independent content review is not recorded.

See language coverage and data snapshot →