mlsanityData to 13 Oct 2026

Are Rules Meant to be Broken? Understanding Multilingual Moral Reasoning as a Computational Pipeline with UniMoral

arXiv 2502.14083 ↗First posted 19 Feb 2025cs.CLcited by 14

On the map: LLM reasoning → Reasoning benchmarks and evaluation

Abstract (arXiv)

Moral reasoning is a complex cognitive process shaped by individual experiences and cultural contexts and presents unique challenges for computational analysis. While natural language processing (NLP) offers promising tools for studying this phenomenon, current research lacks cohesion, employing discordant datasets and tasks that examine isolated aspects of moral reasoning. We bridge this gap with UniMoral, a unified dataset integrating psychologically grounded and social-media-derived moral dilemmas annotated with labels for action choices, ethical principles, contributing factors, and consequences, alongside annotators' moral and cultural profiles. Recognizing the cultural relativity of moral reasoning, UniMoral spans six languages, Arabic, Chinese, English, Hindi, Russian, and Spanish, capturing diverse socio-cultural contexts. We demonstrate UniMoral's utility through a benchmark evaluations of three large language models (LLMs) across four tasks: action prediction, moral typology classification, factor attribution analysis, and consequence generation. Key findings reveal that while implicitly embedded moral contexts enhance the moral reasoning capability of LLMs, there remains a critical need for increasingly specialized approaches to further advance moral reasoning in these models.

Key facts. Cited by 14 papers in the map (data through 13 Oct 2026). Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning. Builds on 22 of its 85 references that are in the map. First posted on arXiv 19 Feb 2025.

Builds on

22 of 85 references.

References · 22

  1. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,223
  2. 2019BERTScore: Evaluating Text Generation with BERTcited by 5,480
  3. 2024Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phonecited by 1,908
  4. 2020Aligning AI With Shared Human Valuescited by 717
  5. 2023Large Language Models Empowered Agent-based Modeling and Simulation: A Survey and Perspectivescited by 255
  6. 2020Scruples: A Corpus of Community Ethical Judgments on 32,000 Real-Life Anecdotescited by 101
  7. 2022When to Make Exceptions: Exploring Language Models as Accounts of Human Moral Judgmentcited by 93
  8. 2022POTATO: The Portable Text Annotation Toolcited by 64
  9. 2022Aligning to Social Norms and Values in Interactive Narrativescited by 32
  10. 2020Moral Framing and Ideological Bias of Newscited by 29
  11. 2021Identifying Morality Frames in Political Tweets using Relational Learningcited by 26
  12. 2024Social Science Meets LLMs: How Reliable Are Large Language Models in Social Simulations?cited by 26
10 more
  1. 2022The Moral Debater: A Study on the Computational Generation of Morally Framed Argumentscited by 22
  2. 2019BERT has a Moral Compass: Improvements of ethical and moral values of machinescited by 21
  3. 2022A Corpus for Understanding and Generating Moral Storiescited by 16
  4. 2024Real or Robotic? Assessing Whether LLMs Accurately Simulate Qualities of Human Responses in Dialoguecited by 15
  5. 2023Values, Ethics, Morals? On the Use of Moral Concepts in NLP Researchcited by 13
  6. 2023Hybrid Approaches for Moral Value Alignment in AI Agents: a Manifestocited by 7
  7. 2025LLMs and Childhood Safety: Identifying Risks and Proposing a Protection Framework for Safe Child-LLM Interactioncited by 7
  8. 2018Development and Validation of the Japanese Moral Foundations Dictionarycited by 6
  9. 2022Broadening AI Ethics Narratives: An Indic Art Viewcited by 3
  10. 2023STREAM: Social data and knowledge collective intelligence platform for TRaining Ethical AI Modelscited by 3

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2025: 6; 2026: 8.

14 papers.

Sort:

By area:

  1. 2025Structured Moral Reasoning in Language Models: A Value-Grounded Evaluation Frameworkcited by 16
  2. 2025CLASH: Evaluating Language Models on Judging High-Stakes Dilemmas from Multiple Perspectivescited by 10
  3. 2025Modeling Annotator Disagreement with Demographic-Aware Experts and Synthetic Perspectivescited by 10
  4. 2025Diagnosing Moral Reasoning Acquisition in Language Models: Pragmatics and Generalizationcited by 6
  5. 2025One Model, Many Morals: Uncovering Cross-Linguistic Misalignments in Computational Moral Reasoningcited by 4
  6. 2025Cross-Lingual Prompt Steerability: Towards Accurate and Robust LLM Behavior across Languagescited by 3
  7. 2026Do Emotions Influence Moral Judgment in Large Language Models?cited by 2
  8. 2026Common Sense vs. Morality: The Curious Case of Narrative Focus Bias in LLMscited by 1
  9. 2026Framing Matters: Addressing Framing Sensitivity in Decision-Making through Behaviorally-Grounded Value Alignmentcited by 0
  10. 2026Which Institutional Frameworks Do Chatbots Assume? Auditing Jurisdictional Defaults in Multilingual LLMscited by 0
  11. 2026Moral Safety in LLMs: Exposing Performative Compliance with Puzzled Cuescited by 0
  12. 2026MET: Theory-Grounded and Culture-Aware Multilingual Moral Reasoningcited by 0
2 more
  1. 2026A Unified Moral-Value Dataset for Instruction Tuningcited by 0
  2. 2026CultureMINE: Datasets and Methods for Improving the Cultural Capabilities of NLP Systemscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Untangling Input Language from Reasoning Language: A Diagnostic Framework for Cross-Lingual Moral Alignment in LLMs
  2. 2024Whose Morality Do They Speak? Unraveling Cultural Bias in Multilingual Language Models
  3. 2025"Pull or Not to Pull?'': Investigating Moral Biases in Leading Large Language Models Across Ethical Dilemmas
  4. 2025Moral Reasoning Across Languages: The Critical Role of Low-Resource Languages in LLMs
  5. 2025Cultural Bias in Large Language Models: Evaluating AI Agents through Moral Questionnaires
  6. 2026Unsupervised Elicitation of Moral Values from Language Models
  7. 2025From Training to Generalization: Improving Moral Reasoning Through Pragmatic Inference
  8. 2026Reasoning or Rhetoric? An Empirical Analysis of Moral Reasoning Explanations in Large Language Models