# Cultural Misalignment in Large Language Models: Detection, Measurement, and Mitigation Through Targeted Fine-Tuning

arXiv: https://arxiv.org/abs/2609.04485 · HTML: https://mlsanity.com/paper/2609.04485/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 23 of its 38 references that are in the map.
- First posted on arXiv 3 Sep 2026.

## Builds on (most cited first, 23 of 23)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Whose Opinions Do Language Models Reflect?](https://mlsanity.com/paper/2303.17548/) (2023)
- [Bias and Fairness in Large Language Models: A Survey](https://mlsanity.com/paper/2309.00770/) (2023)
- [Toxicity in ChatGPT: Analyzing Persona-assigned Language Models](https://mlsanity.com/paper/2304.05335/) (2023)
- [Measurement and Fairness](https://mlsanity.com/paper/1912.05511/) (2019)
- [Quantifying the Persona Effect in LLM Simulations](https://mlsanity.com/paper/2402.10811/) (2024)
- [Cultural Bias and Cultural Alignment of Large Language Models](https://mlsanity.com/paper/2311.14096/) (2023)
- [Collective Constitutional AI: Aligning a Language Model with Public Input](https://mlsanity.com/paper/2406.07814/) (2024)
- [Aligning Language Models to User Opinions](https://mlsanity.com/paper/2305.14929/) (2023)
- [WorldValuesBench: A Large-Scale Benchmark Dataset for Multi-Cultural Value Awareness of Language Models](https://mlsanity.com/paper/2404.16308/) (2024)
- [Beyond prompt brittleness: Evaluating the reliability and consistency of political worldviews in LLMs](https://mlsanity.com/paper/2402.17649/) (2024)
- [CultureLLM: Incorporating Cultural Differences into Large Language Models](https://mlsanity.com/paper/2402.10946/) (2024)
- [Random Silicon Sampling: Simulating Human Sub-Population Opinion Using a Large Language Model Based on Group-Level Demographic Information](https://mlsanity.com/paper/2402.18144/) (2024)
- [AI-Augmented Surveys: Leveraging Large Language Models and Surveys for Opinion Prediction](https://mlsanity.com/paper/2305.09620/) (2023)
- [Personas as a Way to Model Truthfulness in Language Models](https://mlsanity.com/paper/2310.18168/) (2023)
- [OpinionGPT: Modelling Explicit Biases in Instruction-Tuned LLMs](https://mlsanity.com/paper/2309.03876/) (2023)
- [Evaluating Psychological Safety of Large Language Models](https://mlsanity.com/paper/2212.10529/) (2022)
- [The Potential and Challenges of Evaluating Attitudes, Opinions, and Values in Large Language Models](https://mlsanity.com/paper/2406.11096/) (2024)
- [Demonstrations of the Potential of AI-based Political Issue Polling](https://mlsanity.com/paper/2307.04781/) (2023)
- [On the Alignment of Large Language Models with Global Human Opinion](https://mlsanity.com/paper/2509.01418/) (2025)
- [Does Cross-Cultural Alignment Change the Commonsense Morality of Language Models?](https://mlsanity.com/paper/2406.16316/) (2024)
- [Revealing Fine-Grained Values and Opinions in Large Language Models](https://mlsanity.com/paper/2406.19238/) (2024)
- [Investigating Political and Demographic Associations in Large Language Models Through Moral Foundations Theory](https://mlsanity.com/paper/2510.13902/) (2025)

## Abstract (arXiv)

We evaluate three open-weight LLMs (Gemma3-12B from the USA, Bielik-11B-v3 from Poland, and Qwen3-4B from China) against World Values Survey Wave 7 data for 63 demographic personas across three countries, using normalized Wasserstein distance to quantify distributional misalignment. Contrary to expectations, no model favors its home country: the Chinese-built Qwen3-4B performs worst on its own Chinese population (W1 = 0.436, the highest misalignment in the entire model x country matrix). Targeted LoRA fine-tuning on the five worst-case personas, requiring fewer than 1,200 training pairs and under 15 minutes on a single GPU, reduces bias by 16.8% for Bielik-11B (p_Bonf = 0.002, d = -4.4) with all five targets improving. However, country-level decomposition reveals that fine-tuning redistributes rather than removes bias: Bielik's worst-case personas swap entirely from American to Chinese elderly, with zero overlap between pre- and post-correction sets. To our knowledge, this is the first study to target worst-case demographic personas with LoRA fine-tuning for cross-cultural bias mitigation.