mlsanityData to 14 Oct 2026

Multilingual Prompting for Improving LLM Generation Diversity

arXiv 2505.15229 ↗First posted 21 May 2025cs.CL cs.CYcited by 25

Not on the map. Why?

Abstract (arXiv)

Large Language Models (LLMs) are known to lack cultural representation and overall diversity in their generations, from expressing opinions to answering factual questions. To mitigate this problem, we propose multilingual prompting: a prompting method which generates several variations of a base prompt with added cultural and linguistic cues from several cultures, generates responses, and then combines the results. Building on evidence that LLMs have language-specific knowledge, multilingual prompting seeks to increase diversity by activating a broader range of cultural knowledge embedded in model training data. Through experiments across multiple models (GPT-4o, GPT-4o-mini, LLaMA 70B, and LLaMA 8B), we show that multilingual prompting consistently outperforms existing diversity-enhancing techniques such as high-temperature sampling, step-by-step recall, and persona prompting. Further analyses show that the benefits of multilingual prompting vary between high and low resource languages and across model sizes, and that aligning the prompting language with cultural cues reduces hallucination about culturally-specific information.

Key facts. Cited by 25 papers in the map (data through 14 Oct 2026). Builds on 18 of its 36 references that are in the map. First posted on arXiv 21 May 2025.

Builds on

18 of 36 references.

References · 18

  1. 2019HellaSwag: Can a Machine Really Finish Your Sentence?cited by 4,100
  2. 2023Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formattingcited by 817
  3. 2023Whose Opinions Do Language Models Reflect?cited by 773
  4. 2022Language Models are Multilingual Chain-of-Thought Reasonerscited by 530
  5. 2020Social Chemistry 101: Learning to Reason about Social and Moral Normscited by 255
  6. 2023Marked Personas: Using Natural Language Prompts to Measure Stereotypes in Language Modelscited by 207
  7. 2023Does Writing with Language Models Reduce Content Diversity?cited by 192
  8. 2024Large language models that replace human participants can harmfully misportray and flatten identity groupscited by 144
  9. 2020Evaluating the Evaluation of Diversity in Natural Language Generationcited by 129
  10. 2023Increasing Diversity While Maintaining Accuracy: Text Data Generation with Large Language Models and Human Interventionscited by 125
  11. 2024A Note on Shumailov et al. (2024): `AI Models Collapse When Trained on Recursively Generated Data'cited by 115
  12. 2022Picking on the Same Person: Does Algorithmic Monoculture lead to Outcome Homogenization?cited by 91
6 more
  1. 2023Sensitivity, Performance, Robustness: Deconstructing the Effect of Sociodemographic Promptingcited by 81
  2. 2023Improving Diversity of Demographic Representation in Large Language Models via Collective-Critiques and Self-Votingcited by 46
  3. 2024Evaluating Cultural Adaptability of a Large Language Model via Simulation of Synthetic Personascited by 37
  4. 2024Cultural Conditioning or Placebo? On the Effectiveness of Socio-Demographic Promptingcited by 22
  5. 2023Separating form and meaning: Using self-consistency to quantify task understanding across multiple sensescited by 15
  6. 2025Language Models' Factuality Depends on the Language of Inquirycited by 14

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 8; 2026: 17.

25 papers.

Sort:

By area:

  1. 2025The Homogenizing Effect of Large Language Models on Human Expression and Thoughtcited by 35
  2. 2025What and Whose Knowledge? Measuring Epistemic Diversity in Large Language Modelscited by 14
  3. 2025Optimizing Diversity and Quality through Base-Aligned Model Collaborationcited by 8
  4. 2025CrowdLLM: Building LLM-Based Digital Populations Augmented with Generative Modelscited by 5
  5. 2025ConlangCrafter: Constructing Languages with a Multi-Hop LLM Pipelinecited by 4
  6. 2026Language of Thought Shapes Output Diversity in Large Language Modelscited by 4
  7. 2026Steering LLMs for Culturally Localized Generationcited by 4
  8. 2026Addressing LLM Diversity by Infusing Random Conceptscited by 3
  9. 2025Camellia: Benchmarking Cultural Biases in LLMs for Asian Languagescited by 2
  10. 2026No Single Best Model for Diversity: Learning a Router for Sample Diversitycited by 2
  11. 2026Exploring the Capability Boundaries of LLMs in Mastering of Chinese Chouxiang Languagecited by 2
  12. 2026Sampling More, Getting Less: Calibration is the Diversity Bottleneck in LLMscited by 2
13 more
  1. 2025CIEGAD: Cluster-Conditioned Interpolative and Extrapolative Framework for Geometry-Aware and Domain-Aligned Data Augmentationcited by 1
  2. 2026Inference-Time Steering for Cross-Lingual Factual Consistency in LLMscited by 1
  3. 2026Forty Shades of Blue: Quality-Diversity Alignment via Mode-Conditioned Reinforcement Learningcited by 1
  4. 2025PromptGuard at BLP-2025 Task 1: A Few-Shot Classification Framework Using Majority Voting and Keyword Similarity for Bengali Hate Speech Detectioncited by 0
  5. 2026LLM Jaggedness Unlocks Scientific Creativitycited by 0
  6. 2026When English Rewrites Local Knowledge: Global Narrative Dominance in Large Language Modelscited by 0
  7. 2026More Is Not More: What Matters for Diversity in LLM Opinions?cited by 0
  8. 2026Which India Survives Translation? Narrative Homogenisation Across Indian Oral Traditions in LLMscited by 0
  9. 2026Personas Differ from Native-Language Generation: Language Pathways Shape LLM Interpersonal Advicecited by 0
  10. 2026LLMs Mirror Country-Specific Gender Patterns If Asked, but Skew Male When Generating Media in Local Languagescited by 0
  11. 2026Why Pretraining Fails to Share Cross-Lingual Knowledgecited by 0
  12. 2026Improving the Diversity of LLM Outputs without a Trade-offcited by 0
  13. 2026Reach Into The CHOIR: Free-List Elicitation Uncovers Distinct Model Voices in LLM Ensemblescited by 0