mlsanityData to 14 Oct 2026

BengaliMoralBench: A Benchmark for Auditing Moral Reasoning in Large Language Models within Bengali Language and Culture

arXiv 2511.03180 ↗First posted 5 Nov 2025cs.CLcited by 1

On the map: LLM reasoning → Reasoning benchmarks and evaluation

Abstract (arXiv)

As multilingual Large Language Models (LLMs) gain traction across South Asia, their alignment with local ethical norms, particularly for Bengali, spoken by over 285 million people worldwide and among the most widely spoken languages globally, remains underexplored. Existing ethics benchmarks are predominantly English-centric and shaped by Western moral frameworks, overlooking cultural nuances vital for real-world deployment. To address this gap, we introduce BengaliMoralBench, a large-scale ethics benchmark designed for Bengali language and sociocultural contexts. Our benchmark spans five moral domains: (1) Daily Activities, (2) Habits, (3) Parenting, (4) Family Relationships, and (5) Religious Activities, each subdivided into ten culturally grounded categories, totaling 50 subtopics. Each scenario is annotated through native-speaker consensus under three ethical lenses: virtue ethics, commonsense ethics, and justice ethics. We conduct a systematic zero-shot evaluation under a unified prompting protocol across both open-weight and closed-source models, including recent Llama and Gemma variants, Qwen and DeepSeek models, frontier models (GPT-4o-mini and Gemini 1.5 Pro), and a large multilingual baseline (Qwen3-Next-80B). Results show substantial variation in performance across lenses and domains, and our qualitative analysis reveals persistent weaknesses in cultural grounding, commonsense reasoning, and moral fairness. These findings expose critical limitations of current LLMs in non-Western settings and underscore the need for culturally grounded evaluation. BengaliMoralBench provides a foundation for responsible localization and benchmarking to support the deployment of language technologies in culturally diverse, low-resource markets such as Bangladesh.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning. Builds on 33 of its 65 references that are in the map. First posted on arXiv 5 Nov 2025.

Builds on

33 of 65 references.

Closest ancestors · 1

Same direction, newest first

  1. 2025Nunchi-Bench: Benchmarking Language Models on Cultural Reasoning with a Focus on Korean Superstition5 Jul

Other references · 32

  1. 2020Measuring Massive Multitask Language Understandingcited by 7,963
  2. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,720
  3. 2021TruthfulQA: Measuring How Models Mimic Human Falsehoodscited by 2,975
  4. 2020Aligning AI With Shared Human Valuescited by 723
  5. 2022ToxiGen: A Large-Scale Machine-Generated Dataset for Adversarial and Implicit Hate Speech Detectioncited by 422
  6. 2023Cultural Bias and Cultural Alignment of Large Language Modelscited by 149
  7. 2024Continual Learning for Large Language Models: A Surveycited by 147
  8. 2021Re-imagining Algorithmic Fairness in India and Beyondcited by 104
  9. 2024Survey of Cultural Awareness in Language Models: Text and Beyondcited by 103
  10. 2024INCLUDE: Evaluating Multilingual Language Understanding with Regional Knowledgecited by 86
  11. 2024AI Suggestions Homogenize Writing Toward Western Styles and Diminish Cultural Nuancescited by 60
  12. 2025Randomness, Not Representation: The Unreliability of Evaluating Cultural Alignment in LLMscited by 54
20 more
  1. 2024MoralBench: Moral Evaluation of LLMscited by 50
  2. 2024NormAd: A Framework for Measuring the Cultural Adaptability of Large Language Modelscited by 48
  3. 2025The State of Multilingual LLM Safety Research: From Measuring the Language Gap to Mitigating Itcited by 45
  4. 2024Culturally Aware and Adapted NLP: A Taxonomy and a Survey of the State of the Artcited by 26
  5. 2025LLM Ethics Benchmark: A Three-Dimensional Assessment System for Evaluating Moral Reasoning in Large Language Modelscited by 21
  6. 2024The "Colonial Impulse" of Natural Language Processing: An Audit of Bengali Sentiment Analysis Tools and Their Identity-based Biasescited by 16
  7. 2025TigerLLM - A Family of Bangla Large Language Modelscited by 16
  8. 2025SweEval: Do LLMs Really Swear? A Safety Benchmark for Testing Limits for Enterprise Usecited by 15
  9. 2025How Deep Is Representational Bias in LLMs? The Cases of Caste and Religioncited by 14
  10. 2023Decolonial AI Alignment: Openness, Vi'sedsa-Dharma, and Including Excluded Knowledgescited by 10
  11. 2025The Multilingual Divide and Its Impact on Global AI Safetycited by 10
  12. 2025Do You Know About My Nation? Investigating Multilingual Language Models' Cultural Literacy Through Factual Knowledgecited by 8
  13. 2025BnMMLU: Measuring Massive Multitask Language Understanding in Bengalicited by 7
  14. 2025Kaleidoscope: In-language Exams for Massively Multilingual Vision Evaluationcited by 6
  15. 2025We Politely Insist: Your LLM Must Learn the Persian Art of Taarofcited by 6
  16. 2025WorldView-Bench: A Benchmark for Evaluating Global Cultural Perspectives in Large Language Modelscited by 5
  17. 2024Exploring Bengali Religious Dialect Biases in Large Language Models with Evaluation Perspectivescited by 4
  18. 2025BLUCK: A Benchmark Dataset for Bengali Linguistic Understanding and Cultural Knowledgecited by 3
  19. 2021Artificial Intelligence in the Global South (AI4D): Potential and Riskscited by 2
  20. 2025Preserving Cultural Identity with Context-Aware Translation Through Multi-Agent AI Systemscited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Register Shifts Break LLM Safety: A Bengali Benchmark with Culturally Grounded Harmscited by 1

Similar, not cited

Close in wording, no citation either way

  1. 2025One Model, Many Morals: Uncovering Cross-Linguistic Misalignments in Computational Moral Reasoning
  2. 2024Ethical Reasoning and Moral Value Alignment of LLMs Depend on the Language we Prompt them in
  3. 2026MET: Theory-Grounded and Culture-Aware Multilingual Moral Reasoning
  4. 2024Whose Morality Do They Speak? Unraveling Cultural Bias in Multilingual Language Models
  5. 2024Exploring and steering the moral compass of Large Language Models
  6. 2025EvalMORAAL: Interpretable Chain-of-Thought and LLM-as-Judge Evaluation for Moral Alignment in Large Language Models
  7. 2025Are Rules Meant to be Broken? Understanding Multilingual Moral Reasoning as a Computational Pipeline with UniMoral
  8. 2025Structured Moral Reasoning in Language Models: A Value-Grounded Evaluation Framework