Mitigating Cultural Bias in LLMs via Multi-Agent Cultural Debate

• URL: http://arxiv.org/abs/2601.12091v1
• Date: Sat, 17 Jan 2026 16:00:34 GMT
• Title: Mitigating Cultural Bias in LLMs via Multi-Agent Cultural Debate
• Authors: Qian Tan, Lei Jiang, Yuting Zeng, Shuoyang Ding, Xiaohua Xu,
• Abstract summary: Large language models (LLMs) exhibit systematic Western-centric bias, yet whether prompting in non-Western languages can mitigate this remains understudied.<n>We introduce CEBiasBench, a Chinese--English bilingual benchmark, and Multi-Agent Vote (MAV), which enables explicit no bias'' judgments.<n>Using this framework, we find that Chinese prompting merely shifts bias toward East Asian perspectives rather than eliminating it.
• Score: 7.517766226036547
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large language models (LLMs) exhibit systematic Western-centric bias, yet whether prompting in non-Western languages (e.g., Chinese) can mitigate this remains understudied. Answering this question requires rigorous evaluation and effective mitigation, but existing approaches fall short on both fronts: evaluation methods force outputs into predefined cultural categories without a neutral option, while mitigation relies on expensive multi-cultural corpora or agent frameworks that use functional roles (e.g., Planner--Critique) lacking explicit cultural representation. To address these gaps, we introduce CEBiasBench, a Chinese--English bilingual benchmark, and Multi-Agent Vote (MAV), which enables explicit ``no bias'' judgments. Using this framework, we find that Chinese prompting merely shifts bias toward East Asian perspectives rather than eliminating it. To mitigate such persistent bias, we propose Multi-Agent Cultural Debate (MACD), a training-free framework that assigns agents distinct cultural personas and orchestrates deliberation via a "Seeking Common Ground while Reserving Differences" strategy. Experiments demonstrate that MACD achieves 57.6% average No Bias Rate evaluated by LLM-as-judge and 86.0% evaluated by MAV (vs. 47.6% and 69.0% baseline using GPT-4o as backbone) on CEBiasBench and generalizes to the Arabic CAMeL benchmark, confirming that explicit cultural representation in agent frameworks is essential for cross-cultural fairness.

Related papers

• I Am Aligned, But With Whom? MENA Values Benchmark for Evaluating Cultural Alignment and Multilingual Bias in LLMs [5.060243371992739]
  We introduce MENAValues, a novel benchmark designed to evaluate the cultural alignment and multilingual biases of large language models (LLMs)<n> Drawing from large-scale, authoritative human surveys, we curate a structured dataset that captures the sociocultural landscape of MENA with population-level response distributions from 16 countries.<n>Our analysis reveals three critical phenomena: "Cross-Lingual Value Shifts" where identical questions yield drastically different responses based on language, "Reasoning-Induced Degradation" where prompting models to explain their reasoning worsens cultural alignment, and "Logit Leakage" where models refuse sensitive questions while internal probabilities reveal strong hidden
  arXiv  Detail & Related papers  (2025-10-15T05:10:57Z)
• MMA-ASIA: A Multilingual and Multimodal Alignment Framework for Culturally-Grounded Evaluation [91.22008265721952]
  MMA-ASIA centers on a human-curated, multilingual, and multimodally aligned benchmark covering 8 Asian countries and 10 languages.<n>This is the first dataset aligned at the input level across three modalities: text, image (visual question answering), and speech.<n>We propose a five-dimensional evaluation protocol that measures: (i) cultural-awareness disparities across countries, (ii) cross-lingual consistency, (iii) cross-modal consistency, (iv) cultural knowledge generalization, and (v) grounding validity.
  arXiv  Detail & Related papers  (2025-10-07T14:12:12Z)
• CCD-Bench: Probing Cultural Conflict in Large Language Model Decision-Making [0.9310318514564272]
  Large language models can navigate explicit conflicts between legitimately different cultural value systems.<n>CCD-Bench is a benchmark that assesses decision-making under cross-cultural value conflict.<n>CCD-Bench shifts evaluation beyond isolated bias detection toward pluralistic decision making.
  arXiv  Detail & Related papers  (2025-10-03T22:55:37Z)
• Which Cultural Lens Do Models Adopt? On Cultural Positioning Bias and Agentic Mitigation in LLMs [53.07843733899881]
  Large language models (LLMs) have unlocked a wide range of downstream generative applications.<n>We find that they also risk perpetuating subtle fairness issues tied to culture, positioning their generations from the perspectives of the mainstream US culture.<n>We propose 2 inference-time mitigation methods to resolve these biases.
  arXiv  Detail & Related papers  (2025-09-25T12:28:25Z)
• Cross-Cultural Transfer of Commonsense Reasoning in LLMs: Evidence from the Arab World [68.19795061447044]
  This paper investigates cross-cultural transfer of commonsense reasoning in the Arab world.<n>Using a culturally grounded commonsense reasoning dataset covering 13 Arab countries, we evaluate lightweight alignment methods.<n>Our results show that merely 12 culture-specific examples from one country can improve performance in others by 10% on average.
  arXiv  Detail & Related papers  (2025-09-23T17:24:14Z)
• MyCulture: Exploring Malaysia's Diverse Culture under Low-Resource Language Constraints [7.822567458977689]
  MyCulture is a benchmark designed to comprehensively evaluate Large Language Models (LLMs) on Malaysian culture.<n>Unlike conventional benchmarks, MyCulture employs a novel open-ended multiple-choice question format without predefined options.<n>We analyze structural bias by comparing model performance on structured versus free-form outputs, and assess language bias through multilingual prompt variations.
  arXiv  Detail & Related papers  (2025-08-07T14:17:43Z)
• Multiple LLM Agents Debate for Equitable Cultural Alignment [52.01956042197423]
  We introduce a Multi-Agent Debate framework, where two LLM-based agents debate over a cultural scenario and collaboratively reach a final decision.<n>We evaluate these approaches on 7 open-weight LLMs (and 21 LLM combinations) using the NormAd-ETI benchmark for social etiquette norms in 75 countries.<n>Experiments show that debate improves both overall accuracy and cultural group parity over single-LLM baselines.
  arXiv  Detail & Related papers  (2025-05-30T15:01:52Z)
• Multimodal Cultural Safety: Evaluation Frameworks and Alignment Strategies [58.88053690412802]
  Large vision-language models (LVLMs) are increasingly deployed in globally distributed applications, such as tourism assistants.<n> CROSS is a benchmark designed to assess the cultural safety reasoning capabilities of LVLMs.<n>We evaluate 21 leading LVLMs, including mixture-of-experts models and reasoning models.
  arXiv  Detail & Related papers  (2025-05-20T23:20:38Z)
• WorldView-Bench: A Benchmark for Evaluating Global Cultural Perspectives in Large Language Models [1.094065133109559]
  Large Language Models (LLMs) are predominantly trained and aligned in ways that reinforce Western-centric epistemologies and socio-cultural norms.<n>We introduce WorldView-Bench, a benchmark designed to evaluate Global Cultural Inclusivity (GCI) in LLMs by analyzing their ability to accommodate diverse worldviews.
  arXiv  Detail & Related papers  (2025-05-14T17:43:40Z)
• Global MMLU: Understanding and Addressing Cultural and Linguistic Biases in Multilingual Evaluation [71.59208664920452]
  Cultural biases in multilingual datasets pose significant challenges for their effectiveness as global benchmarks.<n>We show that progress on MMLU depends heavily on learning Western-centric concepts, with 28% of all questions requiring culturally sensitive knowledge.<n>We release Global MMLU, an improved MMLU with evaluation coverage across 42 languages.
  arXiv  Detail & Related papers  (2024-12-04T13:27:09Z)

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.