跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2611.20821 · v2

CuRSE-Bench: A Benchmark for Probing Cultural Blind Spots in Text-to-Image Safety Evaluation

首次提交 2026-11-16 18:05
版本更新 2026-11-17 15:43
雷达观测 2026-11-18 19:43
arXiv ↗PDF ↗

摘要

Text-to-image (T2I) safety evaluation typically assigns each prompt or image a single culture-blind harm label, so content that is harmful only within a particular culture can pass as benign. We introduce CultureCraft, an agentic framework that constructs T2I prompts that appear benign under a general-purpose rubric yet carry a latent harmful meaning in a target culture, and use it to build CuRSE-Bench: 10,709 culture-specific prompts and 2,648 culture-general controls over six regions and ten harm axes, each drawn from a predefined knowledge base of documented harm vectors and traceable to at least two independent sources. Scoring every prompt under six region persona judges and a paired culture-blind judge, we find that the culture-blind judge under-rates culture-specific latent harm while matching the persona judges on the controls. Each persona judge, in turn, chiefly decodes its own region's latent harm under our instrument: a selective cultural blind spot that persists in deployed moderators.

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

仅对比实际观测的元数据,不代表 PDF 正文或实验变化。版本可能不连续;带删除线为移除,下划线为新增。

这两个版本的已观测标题、作者、分类、摘要和备注没有变化。正文是否变化请核对原文。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v2 · 2026-11-17 15:43 · 最新观测版本

CuRSE-Bench: A Benchmark for Probing Cultural Blind Spots in Text-to-Image Safety Evaluation

Text-to-image (T2I) safety evaluation typically assigns each prompt or image a single culture-blind harm label, so content that is harmful only within a particular culture can pass as benign. We introduce CultureCraft, an agentic framework that constructs T2I prompts that appear benign under a general-purpose rubric yet carry a latent harmful meaning in a target culture, and use it to build CuRSE-Bench: 10,709 culture-specific prompts and 2,648 culture-general controls over six regions and ten harm axes, each drawn from a predefined knowledge base of documented harm vectors and traceable to at least two independent sources. Scoring every prompt under six region persona judges and a paired culture-blind judge, we find that the culture-blind judge under-rates culture-specific latent harm while matching the persona judges on the controls. Each persona judge, in turn, chiefly decodes its own region's latent harm under our instrument: a selective cultural blind spot that persists in deployed moderators.

查看此版本解读 →查看此版本原文 ↗
v1 · 2026-11-16 18:05

CuRSE-Bench: A Benchmark for Probing Cultural Blind Spots in Text-to-Image Safety Evaluation

Text-to-image (T2I) safety evaluation typically assigns each prompt or image a single culture-blind harm label, so content that is harmful only within a particular culture can pass as benign. We introduce CultureCraft, an agentic framework that constructs T2I prompts that appear benign under a general-purpose rubric yet carry a latent harmful meaning in a target culture, and use it to build CuRSE-Bench: 10,709 culture-specific prompts and 2,648 culture-general controls over six regions and ten harm axes, each drawn from a predefined knowledge base of documented harm vectors and traceable to at least two independent sources. Scoring every prompt under six region persona judges and a paired culture-blind judge, we find that the culture-blind judge under-rates culture-specific latent harm while matching the persona judges on the controls. Each persona judge, in turn, chiefly decodes its own region's latent harm under our instrument: a selective cultural blind spot that persists in deployed moderators.

查看此版本解读 →查看此版本原文 ↗