mlsanityData to 14 Oct 2026

Do Multilingual Language Models Capture Differing Moral Norms?

arXiv 2203.09904 ↗First posted 18 Mar 2022cs.CLcited by 15

Not on the map. Why?

Abstract (arXiv)

Massively multilingual sentence representations are trained on large corpora of uncurated data, with a very imbalanced proportion of languages included in the training. This may cause the models to grasp cultural values including moral judgments from the high-resource languages and impose them on the low-resource languages. The lack of data in certain languages can also lead to developing random and thus potentially harmful beliefs. Both these issues can negatively influence zero-shot cross-lingual model transfer and potentially lead to harmful outcomes. Therefore, we aim to (1) detect and quantify these issues by comparing different models in different languages, (2) develop methods for improving undesirable properties of the models. Our initial experiments using the multilingual model XLM-R show that indeed multilingual LMs capture moral norms, even with potentially higher human-agreement than monolingual ones. However, it is not yet clear to what extent these moral norms differ between languages.

Key facts. Cited by 15 papers in the map (data through 14 Oct 2026). Builds on 14 of its 17 references that are in the map. First posted on arXiv 18 Mar 2022.

Builds on

14 of 17 references.

References · 14

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2020Language Models are Few-Shot Learnerscited by 32,419
  3. 2019Unsupervised Cross-lingual Representation Learning at Scalecited by 3,935
  4. 2019Language Models as Knowledge Bases?cited by 1,936
  5. 2020Aligning AI With Shared Human Valuescited by 723
  6. 2020Making Monolingual Sentence Embeddings Multilingual using Knowledge Distillationcited by 593
  7. 2020AraBERT: Transformer-based Model for Arabic Language Understandingcited by 362
  8. 2021Symbolic Knowledge Distillation: from General Language Models to Commonsense Modelscited by 270
  9. 2019Commonsense Knowledge Mining from Pretrained Modelscited by 189
  10. 2021Large Pre-trained Language Models Contain Human-like Biases of What is Right and Wrong to Docited by 178
  11. 2020German's Next Language Modelcited by 144
  12. 2020On the Language Neutrality of Pre-trained Multilingual Representationscited by 85
2 more
  1. 2021RobeCzech: Czech RoBERTa, a monolingual contextualized language representation modelcited by 33
  2. 2022Combining Static and Contextualised Multilingual Embeddingscited by 8

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 1; 2023: 3; 2024: 6; 2025: 3; 2026: 2.

15 papers.

Sort:

By area:

  1. 2024Emergent social conventions and collective bias in LLM populationscited by 123
  2. 2023Knowledge of cultural moral norms in large language modelscited by 51
  3. 2024Multilingual Large Language Model: A Survey of Resources, Taxonomy and Frontierscited by 51
  4. 2024Emergence of Social Norms in Generative Agent Societies: Principles and Architecturecited by 41
  5. 2022Does Moral Code Have a Moral Code? Probing Delphi's Moral Philosophycited by 31
  6. 2023Values, Ethics, Morals? On the Use of Moral Concepts in NLP Researchcited by 14
  7. 2023Bridging the Language Gap: Dynamic Learning Strategies for Improving Multilingual Performance in LLMscited by 8
  8. 2024Can Agents Spontaneously Form a Society? Introducing a Novel Architecture for Generative Multi-Agents to Elicit Social Emergencecited by 5
  9. 2024Large Language Models as Mirrors of Societal Moral Standardscited by 5
  10. 2024Monolingual and Multilingual Misinformation Detection for Low-Resource Languages: A Comprehensive Surveycited by 4
  11. 2025Exploring Cultural Variations in Moral Judgments with Large Language Modelscited by 3
  12. 2025Visual moral inference and communicationcited by 0
3 more
  1. 2025Probabilistic Aggregation and Targeted Embedding Optimization for Collective Moral Reasoning in Large Language Modelscited by 0
  2. 2026Lost in the Tower of Babel: The Adverse Effects of Incidental Multilingualism in LLMscited by 0
  3. 2026PAVE: A Cognitive Architecture for Legitimate Violation in Generative Agent Societiescited by 0