KSAFE-MM: A Multimodal Safety Benchmark via Localized Contextualization for Korean Cultural Risks

• URL: http://arxiv.org/abs/2605.28013v1
• Date: Wed, 27 May 2026 06:08:54 GMT
• Title: KSAFE-MM: A Multimodal Safety Benchmark via Localized Contextualization for Korean Cultural Risks
• Abstract summary: This paper introduces KSAFE-MM, a benchmark for Korean multimodal safety evaluation.<n>KSAFE-MM consists of two parts, KSAFE-MM-G and KSAFE-MM-C.<n>We evaluate 12 state-of-the-art MLLMs on KSAFE-MM and reveal that models exhibit greater vulnerability to culturally grounded attacks than to generic ones.
• Score: 11.227509564873621
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Multimodal Large Language Models (MLLMs) exacerbate safety risks by introducing vulnerabilities across multiple modalities, such as language and vision. Current MLLM safety evaluation tools, however, suffer from major limitations: 1) English-centric dataset construction, and 2) a focus on generic risks that are not tied to local cultural contexts. This paper introduces KSAFE-MM, a benchmark for Korean multimodal safety evaluation that covers both general safety risks and culture-specific vulnerabilities. KSAFE-MM consists of two parts, KSAFE-MM-G and KSAFE-MM-C. KSAFE-MM-G evaluates globally shared risks in Korean contexts through linguistic contextualization, which transforms generic safety queries into contextually grounded multimodal samples. KSAFE-MM-C targets culture-dependent MLLM safety vulnerabilities using localized visual queries derived from real-world contexts. It pairs these visual queries with jailbreak-style textual queries to cover multimodal safety risks involving cultural visual cues and malicious textual intent. Together, these components provide a general-to-local construction pipeline for evaluating both globally shared safety risks and culture-specific vulnerabilities. We evaluate 12 state-of-the-art MLLMs on KSAFE-MM and reveal that models exhibit greater vulnerability to culturally grounded attacks than to generic ones. Notably, jailbreaking strategies substantially amplify attack success rates, with ProgramExecution yielding up to 74.2% ASR compared to 13.4% for standard queries. Furthermore, we identify a systematic trade-off between safety and over-refusal, where models achieving low ASR tend to exhibit excessive refusal behavior on benign queries. These findings highlight the urgent need for culturally grounded safety evaluation beyond English-centric benchmarks.

Related papers

• SurakshaEval: An Indic Safety Benchmark for Multilingual LLMs [56.34354368047577]
  Existing safety evaluation datasets for large language models (LLMs) predominantly focus on English and Western contexts.<n>We introduce SurakshaEval, a novel safety benchmark composed of human-written prompts spanning real-world scenarios.<n>We benchmark a broad range of state-of-the-art LLMs on SurakshaEval, establish baseline safety performance, and identify recurring failure modes.
  arXiv  Detail & Related papers  (2026-08-08T02:03:33Z)
• XL-SafetyBench: A Country-Grounded Cross-Cultural Benchmark for LLM Safety and Cultural Sensitivity [3.342964361434013]
  XL-SafetyBench is a suite of 5,500 test cases across 10 country-language pairs.<n>It consists of a Jailbreak Benchmark of country-grounded adversarial prompts and a Cultural Benchmark where local sensitivities are embedded within innocuous requests.<n>Each item is constructed via a multi-stage pipeline that combines LLM-assisted principled discovery, automated validation gates, and dual independent native-speaker annotators per country.
  arXiv  Detail & Related papers  (2026-05-07T04:35:03Z)
• SEA-SafeguardBench: Evaluating AI Safety in SEA Languages and Cultures [36.95168918567729]
  Existing multilingual safety benchmarks often rely on machine-translated English data.<n>We introduce SEA-SafeguardBench, the first human-verified safety benchmark for SEA.<n>It covers eight languages, 21,640 samples, across three subsets: general, in-the-wild, and content generation.
  arXiv  Detail & Related papers  (2025-12-05T07:57:57Z)
• OutSafe-Bench: A Benchmark for Multimodal Offensive Content Detection in Large Language Models [54.80460603255789]
  We introduce OutSafe-Bench, the first most comprehensive content safety evaluation test suite designed for the multimodal era.<n>OutSafe-Bench includes a large-scale dataset that spans four modalities, featuring over 18,000 bilingual (Chinese and English) text prompts, 4,500 images, 450 audio clips and 450 videos, all systematically annotated across nine critical content risk categories.<n>In addition to the dataset, we introduce a Multidimensional Cross Risk Score (MCRS), a novel metric designed to model and assess overlapping and correlated content risks across different categories.
  arXiv  Detail & Related papers  (2025-11-13T13:18:27Z)
• When Safe Unimodal Inputs Collide: Optimizing Reasoning Chains for Cross-Modal Safety in Multimodal Large Language Models [50.66979825532277]
  We introduce Safe-Semantics-but-Unsafe-Interpretation (SSUI), the first dataset featuring interpretable reasoning paths tailored for a cross-modal challenge.<n>A novel training framework, Safety-aware Reasoning Path Optimization (SRPO), is also designed based on the SSUI dataset.<n> Experimental results show that our SRPO-trained models achieve state-of-the-art results on key safety benchmarks.
  arXiv  Detail & Related papers  (2025-09-15T15:40:58Z)
• USB: A Comprehensive and Unified Safety Evaluation Benchmark for Multimodal Large Language Models [31.412080488801507]
  Unified Safety Benchmarks (USB) is one of the most comprehensive evaluation benchmarks in MLLM safety.<n>Our benchmark features high-quality queries, extensive risk categories, comprehensive modal combinations, and encompasses both vulnerability and oversensitivity evaluations.
  arXiv  Detail & Related papers  (2025-05-26T08:39:14Z)
• DREAM: Disentangling Risks to Enhance Safety Alignment in Multimodal Large Language Models [37.104276926258095]
  Multimodal Large Language Models (MLLMs) pose unique safety challenges due to their integration of visual and textual data.<n>We introduce textbfDREAM (textittextbfDisentangling textbfRisks to textbfEnhance Safety textbfAlignment in textbfMLLMs), a novel approach that enhances safety alignment in MLLMs through supervised fine-tuning and iterative Reinforcement Learning from AI Feedback.
  arXiv  Detail & Related papers  (2025-04-25T03:54:24Z)
• SafeBench: A Safety Evaluation Framework for Multimodal Large Language Models [75.67623347512368]
  We propose toolns, a comprehensive framework designed for conducting safety evaluations of MLLMs. Our framework consists of a comprehensive harmful query dataset and an automated evaluation protocol. Based on our framework, we conducted large-scale experiments on 15 widely-used open-source MLLMs and 6 commercial MLLMs.
  arXiv  Detail & Related papers  (2024-10-24T17:14:40Z)
• Multimodal Situational Safety [73.63981779844916]
  We present the first evaluation and analysis of a novel safety challenge termed Multimodal Situational Safety.<n>For an MLLM to respond safely, whether through language or action, it often needs to assess the safety implications of a language query within its corresponding visual context.<n>We develop the Multimodal Situational Safety benchmark (MSSBench) to assess the situational safety performance of current MLLMs.
  arXiv  Detail & Related papers  (2024-10-08T16:16:07Z)
• Safety Assessment of Chinese Large Language Models [51.83369778259149]
  Large language models (LLMs) may generate insulting and discriminatory content, reflect incorrect social values, and may be used for malicious purposes. To promote the deployment of safe, responsible, and ethical AI, we release SafetyPrompts including 100k augmented prompts and responses by LLMs.
  arXiv  Detail & Related papers  (2023-04-20T16:27:35Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.