← Back to all news
arXiv AI

Prompt Revision as a Source of Cultural Bias in Text-to-Image Systems

The paper "Prompt Revision as a Source of Cultural Bias in Text-to-Image Systems" investigates how commercial text‑to‑image models silently modify user prompts before generating images, a step that is often hidden from users. Using the multilingual benchmark WORLDVIEW, the authors audit the revision layer in DALL‑E‑3, Imagen‑4, and GPT‑Image‑1.5, finding that non‑Western and non‑Anglophone contexts are disproportionately marked, reduced to narrow vocabularies, and stereotyped. The study demonstrates that the revision layer itself is a previously undocumented causal source of cultural stereotyping, underscoring the need to audit deployed systems rather than just the underlying models.

Related stories

arXiv AI
Oct 27

Beyond Words Alone: Disentangling Distributed Evidence for Visually Grounded Cultural Translation

The paper introduces Culture-MMT, a benchmark for evaluating how well machine translation systems can reconstruct meaning from distributed evidence in social‑media posts. It contains 498 curated, de‑identified user‑generated content instances across 20 domains, classified into Symbol, Visual, and Hybrid based on the type of evidence required. Experiments with 17 model configurations show that real images significantly improve translation for Visual and Hybrid cases, while random images do not, underscoring the importance of accurate visual grounding and cultural adaptation.

By Dingming Li, Yiwen Qiu, Linjuan Wu, Hongxing Li, Zixuan Wang, Xinze Lyu, Ye Guo, Daoxin Zhang, Zhe Xu, Weiming Lu, Yongliang Shen
benchmarks
More like this →
arXiv Computation and Language
Oct 28

The State and Fate of Multilingual, Contextual Evaluation in the NLP World

The paper audits 51 recent multilingual evaluation benchmarks covering 242 datasets and 219 languages, focusing on coverage, representativeness, and rigor. It finds that while coverage is broad, many languages appear in only one benchmark and entire regions are underrepresented; low‑resource languages face a task equity gap. Representativeness is compromised by heavy reliance on English translations, and rigor is weakened by benchmark contamination and parallel‑corpus overlap. The authors propose concrete recommendations for building evaluation frameworks that are natively constructed, culturally grounded, contamination‑aware, and community‑aligned.

By Manan Uppadhyay, Himanshu Beniwal, Prashant Kodali, Sunayana Sitaram
llmsnlpbenchmarks
More like this →
Hugging Face Trending Papers
Oct 27

The State and Fate of Multilingual, Contextual Evaluation in the NLP World

The paper audits 51 multilingual evaluation benchmarks covering 242 datasets and 219 languages, focusing on coverage, representativeness, and rigor. It finds that while coverage is broad, many languages appear in only one benchmark and entire regions are underrepresented, with low-resource languages evaluated on far fewer tasks than high-resource ones. The study also highlights that most datasets rely on English translations, creating artifacts and limiting cultural relevance, and that benchmark contamination through translation leakage and corpus overlap undermines rigor. Recommendations call for natively constructed, culturally grounded, contamination-aware evaluation frameworks that truly serve the communities they aim to assess.

llmsnlpbenchmarks
More like this →
arXiv AI
Aug 25

Beyond Surface Cues: Disentangling Sociocultural Signals in Multilingual LLMs

The study examines how multilingual large language models (LLMs) produce outputs that differ across sociocultural contexts, highlighting that identity labels and source-language cues can mislead assessments of cultural grounding. Using a human‑validated, multi‑agent audit on 89,253 outputs from 12 LLMs in English, French, and Chinese across 18 occupations and three task conditions, the authors find that bias representation varies systematically by language and task. Removing direct identity cues reduces identity‑label prediction in English and Chinese but not in French, and the source language’s cultural context consistently receives the highest relevance score, though this signal weakens after translation or name masking. "whyItMatters":"The findings show that surface cues can obscure true cross‑cultural patterns, underscoring the need for careful audit designs to avoid misleading conclusions about bias in multilingual LLMs."

By Yuanjun Feng, Tanzhou Liu, Stefan Feuerriegel, Yash Raj Shrestha
llmsagentssafety
More like this →
arXiv Computation and Language
Oct 19

Cultural (In)Competence: Investigating Culture-Specific Stereotype Detection Capabilities of LLMs

arXiv:2610.22545v1 Announce Type: new Abstract: Stereotypes are subjective constructs that are mediated by the cultural context in which they are viewed. Successfully modeling them thus also requires...

By Maximilian Splieth\"over, Jannes L\"otfering, Dominik Bachmann, Alexander Espig, Willem Zuidema, Katrin Schulz, Henning Wachsmuth
llmsfine-tuning
More like this →
arXiv AI
Oct 5

CuBEs: Culturally-Situated Behavioral Evaluations and the Limitations of Culture-Blind LLM Judges

CuBEs introduces culturally‑situated behavioral evaluations for large language models, adding cultural context to test scenarios and assessments. The authors built a human‑labeled dataset covering 12 cultures, revealing significant cross‑cultural differences that one‑size‑fits‑all judgments miss. Evaluating 13 LLMs shows that culturally situated tests uncover varied behaviors, such as Western political bias versus non‑Western religious or colonial biases, which standard evaluations overlook.

By Hoda Ayad, Tanu Mitra, Abhishek Mukherji
llmssafety
More like this →