# Are Large Language Models Consistent over Value-laden Questions?

arXiv: https://arxiv.org/abs/2407.02996 · HTML: https://mlsanity.com/paper/2407.02996/

## Key facts

- Cited by 58 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 33 references that are in the map.
- First posted on arXiv 3 Jul 2024.

## Builds on (most cited first, 19 of 19)

- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [Constitutional AI: Harmlessness from AI Feedback](https://mlsanity.com/paper/2212.08073/) (2022)
- [Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2307.15217/) (2023)
- [Towards Measuring the Representation of Subjective Global Opinions in Language Models](https://mlsanity.com/paper/2306.16388/) (2023)
- [Assessing Cross-Cultural Alignment between ChatGPT and Human Societies: An Empirical Study](https://mlsanity.com/paper/2303.17466/) (2023)
- [Fine-tuning language models to find agreement among humans with diverse preferences](https://mlsanity.com/paper/2211.15006/) (2022)
- [Understanding Social Reasoning in Language Models with Language Models](https://mlsanity.com/paper/2306.15448/) (2023)
- [Towards Measuring and Modeling "Culture" in LLMs: A Survey](https://mlsanity.com/paper/2403.15412/) (2024)
- [Probing Pre-Trained Language Models for Cross-Cultural Differences in Values](https://mlsanity.com/paper/2203.13722/) (2022)
- [Language Models as Agent Models](https://mlsanity.com/paper/2212.01681/) (2022)
- [When the Majority is Wrong: Modeling Annotator Disagreement for Subjective Tasks](https://mlsanity.com/paper/2305.06626/) (2023)
- [Do Language Models Have Beliefs? Methods for Detecting, Updating, and Visualizing Model Beliefs](https://mlsanity.com/paper/2111.13654/) (2021)
- [Auxiliary task demands mask the capabilities of smaller language models](https://mlsanity.com/paper/2404.02418/) (2024)
- [Bias-Augmented Consistency Training Reduces Biased Reasoning in Chain-of-Thought](https://mlsanity.com/paper/2403.05518/) (2024)
- [Assessing LLMs for Moral Value Pluralism](https://mlsanity.com/paper/2312.10075/) (2023)
- [Challenging the Validity of Personality Tests for Large Language Models](https://mlsanity.com/paper/2311.05297/) (2023)
- [What does ChatGPT return about human values? Exploring value bias in ChatGPT using a descriptive value theory](https://mlsanity.com/paper/2304.03612/) (2023)
- [Despite "super-human" performance, current LLMs are unsuited for decisions about ethics and safety](https://mlsanity.com/paper/2212.06295/) (2022)
- [The Echoes of Multilinguality: Tracing Cultural Value Shifts during LM Fine-tuning](https://mlsanity.com/paper/2405.12744/) (2024)

## Built on by (most cited first, 30 of 58)

- [Expressing stigma and inappropriate responses prevents LLMs from safely replacing mental health providers](https://mlsanity.com/paper/2504.18412/) (2025)
- [Utility Engineering: Analyzing and Controlling Emergent Value Systems in AIs](https://mlsanity.com/paper/2502.08640/) (2025)
- [Large Language Models Reflect the Ideology of their Creators](https://mlsanity.com/paper/2410.18417/) (2024)
- [Values in the Wild: Discovering and Analyzing Values in Real-World Language Model Interactions](https://mlsanity.com/paper/2504.15236/) (2025)
- [The Personality Illusion: Revealing Dissociation Between Self-Reports & Behavior in LLMs](https://mlsanity.com/paper/2509.03730/) (2025)
- [This human study did not involve human subjects: Validating LLM simulations as behavioral evidence](https://mlsanity.com/paper/2602.15785/) (2026)
- [IssueBench: Millions of Realistic Prompts for Measuring Issue Bias in LLM Writing Assistance](https://mlsanity.com/paper/2502.08395/) (2025)
- [Will AI Tell Lies to Save Sick Children? Litmus-Testing AI Values Prioritization with AIRiskDilemmas](https://mlsanity.com/paper/2505.14633/) (2025)
- [What and Whose Knowledge? Measuring Epistemic Diversity in Large Language Models](https://mlsanity.com/paper/2510.04226/) (2025)
- [Measuring Free-Form Decision-Making Inconsistency of Language Models in Military Crisis Simulations](https://mlsanity.com/paper/2410.13204/) (2024)
- [Prompt Stability Matters: Evaluating and Optimizing Auto-Generated Prompt in General-Purpose Systems](https://mlsanity.com/paper/2505.13546/) (2025)
- [Value Drifts: Tracing Value Alignment During LLM Post-Training](https://mlsanity.com/paper/2510.26707/) (2025)
- [Improving Language Model Personas via Rationalization with Psychological Scaffolds](https://mlsanity.com/paper/2504.17993/) (2025)
- [Large Language Models Often Say One Thing and Do Another](https://mlsanity.com/paper/2503.07003/) (2025)
- [Revisiting LLM Value Probing Strategies: Are They Robust and Expressive?](https://mlsanity.com/paper/2507.13490/) (2025)
- [CoBRA: Programming Cognitive Bias in Social Agents Using Classic Social Science Experiments](https://mlsanity.com/paper/2509.13588/) (2025)
- [The Pluralistic Moral Gap: Understanding Judgment and Value Differences between Humans and Large Language Models](https://mlsanity.com/paper/2507.17216/) (2025)
- [Knowing But Not Doing: Convergent Morality and Divergent Action in LLMs](https://mlsanity.com/paper/2601.07972/) (2026)
- [Large Language Models for Interpretable Mental Health Diagnosis](https://mlsanity.com/paper/2501.07653/) (2025)
- [Evaluate Bias without Manual Test Sets: A Concept Representation Perspective for LLMs](https://mlsanity.com/paper/2505.15524/) (2025)
- [Understanding and Mitigating Political Stance Cross-topic Generalization in Large Language Models](https://mlsanity.com/paper/2508.02360/) (2025)
- [Systematic Bias in Large Language Models: Discrepant Response Patterns in Binary vs. Continuous Judgment Tasks](https://mlsanity.com/paper/2504.19445/) (2025)
- [Evaluating LLM Adaptation to Sociodemographic Factors: User Profile vs. Dialogue History](https://mlsanity.com/paper/2505.21362/) (2025)
- [SocialGaze: Improving the Integration of Human Social Norms in Large Language Models](https://mlsanity.com/paper/2410.08698/) (2024)
- [Generative Psycho-Lexical Approach for Constructing Value Systems in Large Language Models](https://mlsanity.com/paper/2502.02444/) (2025)
- [RankAlign: A Ranking View of the Generator-Validator Gap in Large Language Models](https://mlsanity.com/paper/2504.11381/) (2025)
- [Isolating Culture Neurons in Multilingual Large Language Models](https://mlsanity.com/paper/2508.02241/) (2025)
- [Human Values Matter: Investigating How Misalignment Shapes Collective Behaviors in LLM Agent Communities](https://mlsanity.com/paper/2604.05339/) (2026)
- [Investigating Political and Demographic Associations in Large Language Models Through Moral Foundations Theory](https://mlsanity.com/paper/2510.13902/) (2025)
- [Reward Models Inherit Value Biases from Pretraining](https://mlsanity.com/paper/2601.20838/) (2026)

## Abstract (arXiv)

Large language models (LLMs) appear to bias their survey answers toward certain values. Nonetheless, some argue that LLMs are too inconsistent to simulate particular values. Are they? To answer, we first define value consistency as the similarity of answers across (1) paraphrases of one question, (2) related questions under one topic, (3) multiple-choice and open-ended use-cases of one question, and (4) multilingual translations of a question to English, Chinese, German, and Japanese. We apply these measures to small and large, open LLMs including llama-3, as well as gpt-4o, using 8,000 questions spanning more than 300 topics. Unlike prior work, we find that models are relatively consistent across paraphrases, use-cases, translations, and within a topic. Still, some inconsistencies remain. Models are more consistent on uncontroversial topics (e.g., in the U.S., "Thanksgiving") than on controversial ones ("euthanasia"). Base models are both more consistent compared to fine-tuned models and are uniform in their consistency across topics, while fine-tuned models are more inconsistent about some topics ("euthanasia") than others ("women's rights") like our human subjects (n=165).