mlsanityData to 14 Oct 2026

Mind the Value-Action Gap: Do LLMs Act in Alignment with Their Values?

arXiv 2501.15463 ↗First posted 26 Jan 2025cs.HC cs.AI cs.CLcited by 51

Not on the map. Why?

Abstract (arXiv)

Existing research primarily evaluates the values of LLMs by examining their stated inclinations towards specific values. However, the "Value-Action Gap," a phenomenon rooted in environmental and social psychology, reveals discrepancies between individuals' stated values and their actions in real-world contexts. To what extent do LLMs exhibit a similar gap between their stated values and their actions informed by those values? This study introduces ValueActionLens, an evaluation framework to assess the alignment between LLMs' stated values and their value-informed actions. The framework encompasses the generation of a dataset comprising 14.8k value-informed actions across twelve cultures and eleven social topics, and two tasks to evaluate how well LLMs' stated value inclinations and value-informed actions align across three different alignment measures. Extensive experiments reveal that the alignment between LLMs' stated values and actions is sub-optimal, varying significantly across scenarios and models. Analysis of misaligned results identifies potential harms from certain value-action gaps. To predict the value-action gaps, we also uncover that leveraging reasoned explanations improves performance. These findings underscore the risks of relying solely on the LLMs' stated values to predict their behaviors and emphasize the importance of context-aware evaluations of LLM values and value-action gaps.

Key facts. Cited by 51 papers in the map (data through 14 Oct 2026). Builds on 17 of its 51 references that are in the map. First posted on arXiv 26 Jan 2025.

Builds on

17 of 51 references.

References · 17

  1. 2022Training language models to follow instructions with human feedbackcited by 15,971
  2. 2022Constitutional AI: Harmlessness from AI Feedbackcited by 3,050
  3. 2019ERASER: A Benchmark to Evaluate Rationalized NLP Modelscited by 530
  4. 2020Artificial Intelligence, Values and Alignmentcited by 448
  5. 2024Political Compass or Spinning Arrow? Towards More Meaningful Evaluations for Values and Opinions in Large Language Modelscited by 162
  6. 2024Faithfulness vs. Plausibility: On the (Un)Reliability of Explanations from Large Language Modelscited by 105
  7. 2018Interpretable Deep Learning under Firecited by 74
  8. 2024How Well Do LLMs Represent Values Across Cultures? Empirical Analysis of LLM Responses Based on Hofstede Cultural Dimensionscited by 52
  9. 2024Gender, Race, and Intersectional Bias in Resume Screening via Language Model Retrievalcited by 51
  10. 2024Can Language Models Reason about Individualistic Human Values and Preferences?cited by 34
  11. 2024LLM-GLOBE: A Benchmark Evaluating the Cultural Values Embedded in LLM Outputcited by 27
  12. 2024"They are uncultured": Unveiling Covert Harms and Social Threats in LLM Generated Conversationscited by 25
5 more
  1. 2021A Framework of Severity for Harmful Content Onlinecited by 23
  2. 2024Raising the Bar: Investigating the Values of Large Language Models via Generative Evolving Testingcited by 17
  3. 2022Improving fairness in speaker verification via Group-adapted Fusion Networkcited by 11
  4. 2024Risk and Response in Large Language Models: Evaluating Key Threat Categoriescited by 10
  5. 2024Evaluating Large Language Models with fmevalcited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 15; 2026: 36.

51 papers.

Sort:

By area:

  1. 2025Actions Speak Louder than Words: Agent Decisions Reveal Implicit Biases in Language Modelscited by 30
  2. 2026Whose Facts Win? LLM Source Preferences under Knowledge Conflictscited by 12
  3. 2025Fluent but Foreign: Even Regional LLMs Lack Cultural Alignmentcited by 10
  4. 2025Distributive Fairness in Large Language Models: Evaluating Alignment with Human Valuescited by 9
  5. 2025Do Psychometric Tests Work for Large Language Models? Evaluation of Tests on Sexism, Racism, and Moralitycited by 9
  6. 2026Knowing But Not Doing: Convergent Morality and Divergent Action in LLMscited by 7
  7. 2026Evaluating Alignment of Behavioral Dispositions in LLMscited by 7
  8. 2025Do Role-Playing Agents Practice What They Preach? Belief-Behavior Consistency in LLM-Based Simulations of Human Trustcited by 6
  9. 2025Deep Value Benchmark: Measuring Whether Models Generalize Deep Values or Shallow Preferencescited by 6
  10. 2026The Behavioral Fabric of LLM-Powered GUI Agents: Human Values and Interaction Outcomescited by 4
  11. 2026ValueFlow: Measuring the Propagation of Value Perturbations in Multi-Agent LLM Systemscited by 4
  12. 2026Human Values Matter: Investigating How Misalignment Shapes Collective Behaviors in LLM Agent Communitiescited by 4
38 more
  1. 2026Are Language Models Sensitive to Morally Irrelevant Distractors?cited by 3
  2. 2026The Judgment-Consequence Gap: LLM Moral Reasoning in Healthcare Decisionscited by 3
  3. 2026Affective Context Amplifies Sycophancy in LLM Responsescited by 3
  4. 2025Who Gets the Kidney? Human-AI Alignment, Indecision, and Moral Valuescited by 2
  5. 2025Human Psychometric Questionnaires Mischaracterize LLM Behaviorcited by 2
  6. 2025From Passive to Persuasive: Localized Activation Injection for Empathy and Negotiationcited by 2
  7. 2025Can Finetuing LLMs on Small Human Samples Increase Heterogeneity, Alignment, and Belief-Action Coherence?cited by 2
  8. 2026Lessons Without Borders? Evaluating Cultural Alignment of LLMs Using Multilingual Story Moral Generationcited by 2
  9. 2026To Whom Do Language Models Align? Measuring Principal Hierarchies Under High-Stakes Competing Demandscited by 2
  10. 2026What Does the AI Doctor Value? Auditing Pluralism in the Clinical Ethics of Language Modelscited by 2
  11. 2026ActTraitBench: Quantifying the Knowledge-Decision Gap in Large Language Models via Human-Grounded Behavioral Validationcited by 2
  12. 2025Free Lunch for User Experience: Crowdsourcing Agents for Scalable User Studiescited by 1
  13. 2026Value Alignment Tax: Measuring Value Trade-offs in LLM Alignmentcited by 1
  14. 2026Distributional Open-Ended Evaluation of LLM Cultural Value Alignment Based on Value Codebookcited by 1
  15. 2026Beyond Social Pressure: Benchmarking Epistemic Attack in Large Language Modelscited by 1
  16. 2026Beyond Value Benchmarks: Measuring Value-Structure Alignment in Large Language Models via Symmetric Q-Sortscited by 1
  17. 2026When Preferences Fail to Become Incentives: A Utility-Behavior Gap in Large Language Modelscited by 1
  18. 2026PLURAL: A Global Dataset for Value Alignmentcited by 1
  19. 2026People Are Not Just Their Countries. Disentangling Social Determinants of LLM Value Alignment Across Europecited by 1
  20. 2026STONIC: A Layered Measurement Contract for LLM Value Profilingcited by 1
  21. 2026AI Revealed Preferencescited by 1
  22. 2025Evidence of conceptual mastery in the application of rules by Large Language Modelscited by 0
  23. 2025Conversations: Love Them, Hate Them, Steer Themcited by 0
  24. 2025Offscript: Automated Auditing of Instruction Adherence in LLMscited by 0
  25. 2025Human-AI Interaction Alignment: Designing, Evaluating, and Evolving Value-Centered AI For Reciprocal Human-AI Futurescited by 0
  26. 2026Value-Action Alignment in Large Language Models under Privacy-Prosocial Conflictcited by 0
  27. 2026Can Large Language Models Make Everyone Happy?cited by 0
  28. 2026Are Aligned Large Language Models Still Misaligned?cited by 0
  29. 2026Towards Cross-lingual Values Judgment: A Consensus-Pluralism Perspectivecited by 0
  30. 2026Superficial Beliefs in LLM Decision-Makingcited by 0
  31. 2026Which Values Do LLMs Confuse? A Schwartz-Based Recognition Studycited by 0
  32. 2026A Unified Moral-Value Dataset for Instruction Tuningcited by 0
  33. 2026ExpertIVS: Sociological Expert Driven Individual Value Simulation in Large Language Modelscited by 0
  34. 2026The Failure Happens Before the Drift: The Social Dynamics of Values in LLM Agent Societiescited by 0
  35. 2026CultureMINE: Datasets and Methods for Improving the Cultural Capabilities of NLP Systemscited by 0
  36. 2026Can Open-Weight Large Language Models (LLMs) Simulate Human Survey Populations? A Cross-Instrument Calibration Studycited by 0
  37. 2026On the Behavioral Traits of LLM Agentscited by 0
  38. 2026CuBEs: Culturally-Situated Behavioral Evaluations and the Limitations of Culture-Blind LLM Judgescited by 0

50 most cited of 51 All 51 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026Do LLMs Have Values? A Quantitative Analysis and Alignment Framework for Values in Large Language Models
  2. 2025Do LLMs Align Human Values Regarding Social Biases? Judging and Explaining Social Biases with LLMs
  3. 2025Do Language Models Think Consistently? A Study of Value Preferences Across Varying Response Lengths
  4. 2023Exploring the psychology of LLMs' Moral and Legal Reasoning
  5. 2025Value Drifts: Tracing Value Alignment During LLM Post-Training
  6. 2024DailyDilemmas: Revealing Value Preferences of LLMs with Quandaries of Daily Life
  7. 2023Heterogeneous Value Alignment Evaluation for Large Language Models
  8. 2024Are the Values of LLMs Structurally Aligned with Humans? A Causal Perspective