ACE-Align: Attribute Causal Effect Alignment for Cultural Values under Varying Persona Granularities

• URL: http://arxiv.org/abs/2601.12962v1
• Date: Mon, 19 Jan 2026 11:18:25 GMT
• Title: ACE-Align: Attribute Causal Effect Alignment for Cultural Values under Varying Persona Granularities
• Abstract summary: We propose ACE-Align, a causal-effect framework that aligns how demographic attributes shift different cultural values.<n>Across all persona granularities, ACE-Align consistently outperforms baselines.<n>It improves geographic equity by reducing the average alignment gap between high-resource and low-resource regions from 9.81 to 4.92 points.
• Score: 76.52901967874622
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Ensuring that large language models (LLMs) respect diverse cultural values is crucial for social equity. However, existing approaches often treat cultural groups as homogeneous and overlook within-group heterogeneity induced by intersecting demographic attributes, leading to unstable behavior under varying persona granularity. We propose ACE-Align (Attribute Causal Effect Alignment), a causal-effect framework that aligns how specific demographic attributes shift different cultural values, rather than treating each culture as a homogeneous group. We evaluate ACE-Align across 14 countries spanning five continents, with personas specified by subsets of four attributes (gender, education, residence, and marital status) and granularity instantiated by the number of specified attributes. Across all persona granularities, ACE-Align consistently outperforms baselines. Moreover, it improves geographic equity by reducing the average alignment gap between high-resource and low-resource regions from 9.81 to 4.92 points, while Africa shows the largest average gain (+8.48 points). Code is available at https://github.com/Wells-Luo/ACE-Align.

Related papers

• CultureConverse: A Multilingual Multi-turn Simulation Harness for Culturally Grounded Assistance in East and Southeast Asia [72.20656906184539]
  We introduce CultureConverse, a scalable, multilingual simulation and evaluation harness for culturally grounded assistant dialogue.<n>The resulting CultureConverse-DS dataset contains 14,610 benchmark (evaluation) episodes and 274,295 oracle-guided (gold-mode) dialogues.<n>Human annotation experiments suggest that our evaluation framework is a sufficient proxy for human judgment.
  arXiv  Detail & Related papers  (2026-08-28T14:56:46Z)
• People Are Not Just Their Countries. Disentangling Social Determinants of LLM Value Alignment Across Europe [5.742135983891038]
  Large Language Models (LLMs) are increasingly used as a primary source of information and advice.<n>We consider value alignment displayed with respect to 10 prominent commercial LLMs in terms of 15 socio-demographic variables.<n>Our analyses reveal that LLMs are indeed unequally aligned to the values of different socio-demographic groups.
  arXiv  Detail & Related papers  (2026-08-07T16:10:53Z)
• Whose Norms? Disentangling Cultural and Personal Alignment in Large Language Models [66.34115263472931]
  Large language models are increasingly used for social decision-making situations that require balancing cultural norms with personal preferences.<n>We introduce PACT, the Personal-Preference and Cultural-Norm Trade-off framework, which evaluates whether models choose to follow a cultural norm or allow personal preferences.<n>Our five-country human study on PACT shows that culture-following in humans is mainly driven by scenario country, with the lowest agreement when participants judge their own cultural contexts.
  arXiv  Detail & Related papers  (2026-06-05T22:20:30Z)
• Beyond Universality: The GCC-FER Dataset and Culture-Aware Adaptation for Dynamic Facial Expression Recognition [5.764381280944401]
  We introduce a new hybrid multicultural video dataset termed Global Cross-Cultural Facial Expression Recognition (GCC-FER)<n> GCC-FER comprises 23,934 video samples spanning four cultural groups across seven basic expressions.<n>A Culture-Aware FER (CA-FER) system is proposed to mitigate cultural bias by adaptively recalibrating latent facial representations.
  arXiv  Detail & Related papers  (2026-06-05T09:03:18Z)
• Quantifying the Salience of Geo-Cultural Values for Pluralistic Safety Alignment [8.180786800568125]
  Rater pools in safety evaluation datasets remain largely geographically homogeneous, failing to capture geo-cultural differences.<n>We find that most do not report geo-cultural information, and those that do lack a unified methodology to jointly analyze geo-cultural and demographic correlates.<n>Our findings motivate more culturally pluralistic safety evaluation and offer practical takeaways to support it.
  arXiv  Detail & Related papers  (2026-05-29T21:21:39Z)
• Can Persona-Prompted LLMs Emulate Subgroup Values? An Empirical Analysis of Generalisability and Fairness in Cultural Alignment [75.88231994139132]
  We show that even state-of-the-art models like GPT-4.1 achieve only 57.4% accuracy in predicting modal preferences.<n>We find significant pre-existing performance biases, where models better emulate young, male, Chinese, and Christian personas.
  arXiv  Detail & Related papers  (2026-04-14T15:06:13Z)
• Distributional Open-Ended Evaluation of LLM Cultural Value Alignment Based on Value Codebook [50.10594064510559]
  DOVE is a distributional evaluation framework that directly compares human-written text with LLM-generated outputs.<n>We show that DOVE achieves superior predictive validity, attaining a 31.56% correlation with downstream tasks, while maintaining high reliability with as few as 500 samples per culture.
  arXiv  Detail & Related papers  (2026-03-16T08:33:10Z)
• I Am Aligned, But With Whom? MENA Values Benchmark for Evaluating Cultural Alignment and Multilingual Bias in LLMs [5.060243371992739]
  We introduce MENAValues, a novel benchmark designed to evaluate the cultural alignment and multilingual biases of large language models (LLMs)<n> Drawing from large-scale, authoritative human surveys, we curate a structured dataset that captures the sociocultural landscape of MENA with population-level response distributions from 16 countries.<n>Our analysis reveals three critical phenomena: "Cross-Lingual Value Shifts" where identical questions yield drastically different responses based on language, "Reasoning-Induced Degradation" where prompting models to explain their reasoning worsens cultural alignment, and "Logit Leakage" where models refuse sensitive questions while internal probabilities reveal strong hidden
  arXiv  Detail & Related papers  (2025-10-15T05:10:57Z)
• Cross-Cultural Transfer of Commonsense Reasoning in LLMs: Evidence from the Arab World [68.19795061447044]
  This paper investigates cross-cultural transfer of commonsense reasoning in the Arab world.<n>Using a culturally grounded commonsense reasoning dataset covering 13 Arab countries, we evaluate lightweight alignment methods.<n>Our results show that merely 12 culture-specific examples from one country can improve performance in others by 10% on average.
  arXiv  Detail & Related papers  (2025-09-23T17:24:14Z)
• Cultivating Pluralism In Algorithmic Monoculture: The Community Alignment Dataset [18.197532754060244]
  We show that humans exhibit significantly more variation in preferences than the responses of 21 state-of-the-art LLMs.<n>We argue that this motivates the need for negatively-correlated sampling when generating candidate sets.<n>We collect and open-source Community Alignment, the largest and most representative multilingual and multi-turn preference dataset to date.
  arXiv  Detail & Related papers  (2025-07-13T14:34:22Z)
• CAIRe: Cultural Attribution of Images by Retrieval-Augmented Evaluation [61.130639734982395]
  We introduce CAIRe, a novel evaluation metric that assesses the degree of cultural relevance of an image.<n>Our framework grounds entities and concepts in the image to a knowledge base and uses factual information to give independent graded judgments for each culture label.
  arXiv  Detail & Related papers  (2025-06-10T17:16:23Z)
• CARE: Multilingual Human Preference Learning for Cultural Awareness [48.760262639641496]
  We introduce textbfCARE, a multilingual resource containing 3,490 culturally specific questions and 31.7k responses with human judgments.<n>We demonstrate how a modest amount of high-quality native preferences improves cultural awareness across various LMs.<n>Our analyses reveal that models with stronger initial cultural performance benefit more from alignment.
  arXiv  Detail & Related papers  (2025-04-07T14:57:06Z)
• VITAL: A New Dataset for Benchmarking Pluralistic Alignment in Healthcare [9.087074203425061]
  Existing alignment paradigms fail to account for the diversity of perspectives across cultures, demographics, and communities.<n>This is particularly critical in health-related scenarios, where plurality is essential due to the influence of culture, religion, personal values, and conflicting opinions.<n>This work highlights the limitations of current approaches and lays the groundwork for developing health-specific alignment solutions.
  arXiv  Detail & Related papers  (2025-02-19T14:38:57Z)
• Whose Preferences? Differences in Fairness Preferences and Their Impact on the Fairness of AI Utilizing Human Feedback [8.04095222893591]
  We find significant gaps in fairness preferences depending on the race, age, political stance, educational level, and LGBTQ+ identity of annotators. We also demonstrate that demographics mentioned in text have a strong influence on how users perceive individual fairness in moderation.
  arXiv  Detail & Related papers  (2024-06-09T19:42:25Z)
• D3CODE: Disentangling Disagreements in Data across Cultures on Offensiveness Detection and Evaluation [5.9053106775634685]
  We introduce the dataset: a large-scale cross-cultural dataset of parallel annotations for offensive language in over 4.5K sentences annotated by a pool of over 4k annotators. The dataset contains annotators' moral values captured along six moral foundations: care, equality, proportionality, authority, loyalty, and purity. Our analyses reveal substantial regional variations in annotators' perceptions that are shaped by individual moral values.
  arXiv  Detail & Related papers  (2024-04-16T19:12:03Z)
• GIVL: Improving Geographical Inclusivity of Vision-Language Models with Pre-Training Methods [62.076647211744564]
  We propose GIVL, a Geographically Inclusive Vision-and-Language Pre-trained model. There are two attributes of geo-diverse visual concepts which can help to learn geo-diverse knowledge: 1) concepts under similar categories have unique knowledge and visual characteristics, 2) concepts with similar visual features may fall in completely different categories. Compared with similar-size models pre-trained with similar scale of data, GIVL achieves state-of-the-art (SOTA) and more balanced performance on geo-diverse V&L tasks.
  arXiv  Detail & Related papers  (2023-01-05T03:43:45Z)
• Mitigating Face Recognition Bias via Group Adaptive Classifier [53.15616844833305]
  This work aims to learn a fair face representation, where faces of every group could be more equally represented. Our work is able to mitigate face recognition bias across demographic groups while maintaining the competitive accuracy.
  arXiv  Detail & Related papers  (2020-06-13T06:43:37Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.