mlsanityData to 14 Oct 2026

Cultural Conditioning or Placebo? On the Effectiveness of Socio-Demographic Prompting

arXiv 2406.11661 ↗First posted 17 Jun 2024cs.CLcited by 22

Not on the map. Why?

Abstract (arXiv)

Socio-demographic prompting is a commonly employed approach to study cultural biases in LLMs as well as for aligning models to certain cultures. In this paper, we systematically probe four LLMs (Llama 3, Mistral v0.2, GPT-3.5 Turbo and GPT-4) with prompts that are conditioned on culturally sensitive and non-sensitive cues, on datasets that are supposed to be culturally sensitive (EtiCor and CALI) or neutral (MMLU and ETHICS). We observe that all models except GPT-4 show significant variations in their responses on both kinds of datasets for both kinds of prompts, casting doubt on the robustness of the culturally-conditioned prompting as a method for eliciting cultural bias in models or as an alignment strategy. The work also calls rethinking the control experiment design to tease apart the cultural conditioning of responses from "placebo effect", i.e., random perturbations of model responses due to arbitrary tokens in the prompt.

Key facts. Cited by 22 papers in the map (data through 14 Oct 2026). Builds on 23 of its 42 references that are in the map. First posted on arXiv 17 Jun 2024.

Builds on

23 of 42 references.

References · 23

  1. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
  2. 2023Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formattingcited by 817
  3. 2020Aligning AI With Shared Human Valuescited by 723
  4. 2023Towards Measuring the Representation of Subjective Global Opinions in Language Modelscited by 437
  5. 2023From Pretraining Data to Language Models to Downstream Tasks: Tracking the Trails of Political Biases Leading to Unfair NLP Modelscited by 260
  6. 2023Marked Personas: Using Natural Language Prompts to Measure Stereotypes in Language Modelscited by 207
  7. 2021Large Pre-trained Language Models Contain Human-like Biases of What is Right and Wrong to Docited by 178
  8. 2023Cultural Alignment in Large Language Models: An Explanatory Analysis Based on Hofstede's Cultural Dimensionscited by 112
  9. 2023Aligning Language Models to User Opinionscited by 99
  10. 2023NLPositionality: Characterizing Design Biases of Datasets and Modelscited by 94
  11. 2023SeaEval for Multilingual Foundation Models: From Cross-Lingual Alignment to Cultural Reasoningcited by 94
  12. 2023Sensitivity, Performance, Robustness: Deconstructing the Effect of Sociodemographic Promptingcited by 81
11 more
  1. 2024CULTURE-GEN: Revealing Global Cultural Perception in Language Models through Natural Language Promptingcited by 55
  2. 2023Large Language Models Only Pass Primary School Exams in Indonesia: A Comprehensive Test on IndoMMLUcited by 53
  3. 2023Multi-lingual and Multi-cultural Figurative Language Understandingcited by 47
  4. 2020RiSAWOZ: A Large-Scale Multi-Domain Wizard-of-Oz Dataset with Rich Semantic Annotations for Task-Oriented Dialogue Modelingcited by 41
  5. 2024IndoCulture: Exploring Geographically-Influenced Cultural Commonsense Reasoning Across Eleven Indonesian Provincescited by 39
  6. 2023Are Personalized Stochastic Parrots More Dangerous? Evaluating Persona Biases in Dialogue Systemscited by 37
  7. 2023EtiCor: Corpus for Analyzing LLMs for Etiquettescited by 33
  8. 2022SODAPOP: Open-Ended Discovery of Social Biases in Social Commonsense Reasoning Modelscited by 22
  9. 2023Sociocultural Norm Similarities and Differences via Situational Alignment and Explainable Textual Entailmentcited by 22
  10. 2022Evaluating the Diversity, Equity and Inclusion of NLP Technology: A Case Study for Indian Languagescited by 20
  11. 2023Cultural Compass: Predicting Transfer Learning Success in Offensive Language Detection with Cultural Featurescited by 18

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 4; 2025: 11; 2026: 7.

22 papers.

Sort:

By area:

  1. 2024Survey of Cultural Awareness in Language Models: Text and Beyondcited by 103
  2. 2025Multilingual Prompting for Improving LLM Generation Diversitycited by 25
  3. 2024Which Demographics do LLMs Default to During Annotation?cited by 10
  4. 2025Fluent but Foreign: Even Regional LLMs Lack Cultural Alignmentcited by 10
  5. 2025Culture is Everywhere: A Call for Intentionally Cultural Evaluationcited by 10
  6. 2025Meta-Cultural Competence: Climbing the Right Hill of Cultural Awarenesscited by 8
  7. 2025Revisiting LLM Value Probing Strategies: Are They Robust and Expressive?cited by 8
  8. 2024"A Woman is More Culturally Knowledgeable than A Man?": The Effect of Personas on Cultural Norm Interpretation in LLMscited by 5
  9. 2025Can Third-parties Read Our Emotions?cited by 4
  10. 2025PakBBQ: A Culturally Adapted Bias Benchmark for QAcited by 4
  11. 2024Kalahi: A handcrafted, grassroots cultural LLM evaluation suite for Filipinocited by 3
  12. 2025From Anger to Joy: How Nationality Personas Shape Emotion Attribution in Large Language Modelscited by 2
10 more
  1. 2025XToM: Exploring the Multilingual Theory of Mind for Large Language Modelscited by 2
  2. 2025Improving the Distributional Alignment of LLMs using Supervisioncited by 2
  3. 2025User Behavior Prediction as a Generic, Robust, Scalable, and Low-Cost Evaluation Strategy for Estimating Generalization in LLMscited by 2
  4. 2026Language Models Entangle Language and Culturecited by 1
  5. 2026To Generate or Discriminate? Methodological Considerations for Measuring Cultural Alignment in LLMscited by 0
  6. 2026PartisanLens: A Multilingual Dataset of Hyperpartisan and Conspiratorial Immigration Narratives in European Mediacited by 0
  7. 2026Signal or Spurious Cue? A Randomized Audit of Survey-Country Metadata in LLM Social Inferencecited by 0
  8. 2026Hints Help But Do They Teach? Evaluating Skills Transfer in Code Generationcited by 0
  9. 2026Alignment by Stereotyping: How LLMs Sacrifice Individual Distinctiveness for Cultural Adaptationcited by 0
  10. 2026Measuring Cultural Alignment Beyond the Average: A Framework for Evaluating Maternal-Health LLM Interactions in Indian Contextscited by 0