Going PLACES: Participatory Localized Red Teaming for Text-to-Image Safety in the Global South

• URL: http://arxiv.org/abs/2605.19190v1
• Date: Mon, 18 May 2026 23:34:53 GMT
• Title: Going PLACES: Participatory Localized Red Teaming for Text-to-Image Safety in the Global South
• Abstract summary: Text-to-image (T2I) models are largely calibrated to a Western-centric default, creating significant vulnerabilities for the rest of the world.<n>To embrace cultural pluralism and bring historically under-represented perspectives in T2I safety, we conduct community-centered red teaming studies in the Global South.<n>We present a dataset comprising over 26,000 examples of T2I model failures collected in partnership with universities in Ghana, Nigeria, and two regions of India.
• Score: 17.048182455492896
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Despite the global deployment of text-to-image (T2I) models, their safety frameworks are largely calibrated to a Western-centric default, creating significant vulnerabilities for the rest of the world. To embrace cultural pluralism and bring historically under-represented perspectives in T2I safety, we conduct localised community-centered red teaming studies in the Global South. Our two-fold approach prioritizes localization and participation, by focusing on secondary urban centers in these regions, and conducting community engagement and training workshops to contextualize local norms. As a result, we present PLACES, a dataset comprising over 26,000 examples of T2I model failures collected in partnership with universities in Ghana, Nigeria, and two regions of India (Karnataka and Punjab). Analysis of prompts collected reveals a wide-ranging diversity in socio-cultural and linguistic attributes, when compared to existing geography-agnostic crowdsourced red-teaming data. We observe unique adversarial patterns enabled by local cultural and linguistic nuances, and distinct clusters within region around specific themes, such as religion in India. Moreover, we uncover structural contextual gaps in existing safety frameworks by identifying novel harms showing normative dissonance (e.g., violating religious norms, ignoring local customs, and ominous symbolism). This work argues that expanding T2I safety requires moving beyond mere scale to incorporate deeply localised, participatory methodologies for data collection and contextualization. Content warning: This paper includes examples containing potentially harmful or offensive content.

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)
• Pluralis v0.1: Towards a Multicultural, Multimodal, Multilingual Benchmark for AI Risk and Reliability [46.80780735011334]
  Current AI safety evaluation and benchmarking frameworks rely on Western-centric culture-agnostic defaults that mask critical regional laws, socio-linguistic nuances, and cultural taboos.<n>We introduce Pluralis v0.1: a novel multimodal, multi-regional, and multilingual dataset built from a culture-first perspective.
  arXiv  Detail & Related papers  (2026-07-07T12:21:52Z)
• Mind the Gap: Pitfalls of LLM Alignment with Asian Public Opinion [8.443928474148114]
  Large Language Models (LLMs) are increasingly being deployed in multilingual, multicultural settings.<n>Our study specifically focuses on the sensitive domain of religion as the prism for broader alignment.<n>We find that while the popular models generally align with public opinion on broad social issues, they consistently fail to accurately represent religious viewpoints.
  arXiv  Detail & Related papers  (2026-03-06T13:29:54Z)
• Multilingual Text-to-Image Person Retrieval via Bidirectional Relation Reasoning and Aligning [81.43257201833154]
  We propose Bi-IRRA: a Bidirectional Implicit Relation Reasoning and Aligning framework to learn alignment across languages and modalities.<n>Within Bi-IRRA, a bidirectional implicit relation reasoning module enables bidirectional prediction of masked image and text.<n>The proposed method achieves new state-of-the-art results on all multilingual TIPR datasets.
  arXiv  Detail & Related papers  (2025-10-20T16:01:11Z)
• MINGLE: VLMs for Semantically Complex Region Detection in Urban Scenes [49.89767522399176]
  Group-level social interactions in public spaces are crucial for urban planning.<n>We introduce a social group region detection task, which requires inferring and spatially grounding visual regions defined by interpersonal relations.<n>We propose MINGLE, a modular three-stage pipeline that integrates human detection and depth estimation, VLM-based reasoning to classify pairwise social affiliation, and a lightweight spatial aggregation algorithm to localize socially connected groups.<n>We present a new dataset of 100K urban street-view images annotated with bounding boxes and labels for both individuals and socially interacting groups.
  arXiv  Detail & Related papers  (2025-09-16T19:31:40Z)
• Mind the Gesture: Evaluating AI Sensitivity to Culturally Offensive Non-Verbal Gestures [67.89565097372751]
  Gestures are an integral part of non-verbal communication, with meanings that vary across cultures.<n>As AI systems become more integrated into global applications, ensuring they do not inadvertently perpetuate cultural offenses is critical.<n>We introduce Multi-Cultural Set of Inappropriate Gestures and Nonverbal Signs (MC-SIGNS), a dataset of 288 gesture-country pairs annotated for offensiveness, cultural significance, and contextual factors.
  arXiv  Detail & Related papers  (2025-02-24T23:10:08Z)
• Do Generative AI Models Output Harm while Representing Non-Western Cultures: Evidence from A Community-Centered Approach [8.805524738976073]
  This research investigates the impact of Generative Artificial Intelligence (GAI) models, specifically text-to-image generators (T2Is), on the representation of non-Western cultures.
  arXiv  Detail & Related papers  (2024-07-20T07:01:37Z)
• Against The Achilles' Heel: A Survey on Red Teaming for Generative Models [60.21722603260243]
  Our extensive survey, which examines over 120 papers, introduces a taxonomy of fine-grained attack strategies grounded in the inherent capabilities of language models. We have developed the "searcher" framework to unify various automatic red teaming approaches.
  arXiv  Detail & Related papers  (2024-03-31T09:50:39Z)
• SeeGULL Multilingual: a Dataset of Geo-Culturally Situated Stereotypes [18.991295993710224]
  SeeGULL is a global-scale multilingual dataset of social stereotypes, spanning 20 languages, with human annotations across 23 regions, and demonstrate its utility in identifying gaps in model evaluations.
  arXiv  Detail & Related papers  (2024-03-08T22:09:58Z)
• AI's Regimes of Representation: A Community-centered Study of Text-to-Image Models in South Asia [18.308417975842058]
  We show how generative AI can reproduce an outsiders gaze for viewing South Asian cultures, shaped by global and regional power inequities. We distill lessons for responsible development of T2I models, recommending concrete pathways forward.
  arXiv  Detail & Related papers  (2023-05-19T17:35:20Z)
• Broaden the Vision: Geo-Diverse Visual Commonsense Reasoning [49.04866469947569]
  We construct a Geo-Diverse Visual Commonsense Reasoning dataset (GD-VCR) to test vision-and-language models' ability to understand cultural and geo-location-specific commonsense. We find that the performance of both models for non-Western regions including East Asia, South Asia, and Africa is significantly lower than that for Western region.
  arXiv  Detail & Related papers  (2021-09-14T17:52:55Z)

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.