Metraipolis › archive › dc_4178d56b5ec0
Aligned but Flattened: Analyzing the Trade-off between Cultural Alignment and Diversity in LLMs — arXiv:2609.00565
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.00565
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T06:41:44.323Z
- SHA-256 of the archived text
85372f502501fdffd18267819ef49ed680f88b4e9b6c74fd376a5cf136ae877d- Stable URL of this page
https://metraipolis.com/documento/dc_4178d56b5ec0
How to cite
Aligned but Flattened: Analyzing the Trade-off between Cultural Alignment and Diversity in LLMs — arXiv:2609.00565 — arXiv (en), https://arxiv.org/abs/2609.00565 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4178d56b5ec0 (Metraipolis archive dc_4178d56b5ec0, sha256 85372f502501fdffd18267819ef49ed680f88b4e9b6c74fd376a5cf136ae877d)
BibTeX
@misc{dc4178d56b5ec0,
title = {Aligned but Flattened: Analyzing the Trade-off between Cultural Alignment and Diversity in LLMs — arXiv:2609.00565},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.00565},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4178d56b5ec0; sha256 85372f502501fdffd18267819ef49ed680f88b4e9b6c74fd376a5cf136ae877d},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.00565. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Cultural fine-tuning has become the de facto paradigm for building culture-aware large language models (LLMs), yet existing optimization exclusively for alignment scores provides an incomplete portrait of cultural fidelity by systematically obscuring inherent cultural diversity. This unidimensional evaluation lens prompts a fundamental question: do models genuinely perceive distinct cultural nuances, or do they merely memorize dominant cultural values? To address this, we propose a synergistic evaluation framework that jointly formalizes cultural alignment and diversity. Through extensive benchmarking of six mainstream LLMs on the World Values Survey, this framework uncovers a systematic and critical trade-off: the pursuit of cultural alignment consistently incurs an acute expense of diversity, leading to severe "cultural flattening." Investigating this behavioral shift, we demonstrate that these superficial alignment gains stem from models artificially anchoring to dominant majorities, converging onto a monolithic response pattern that wipes out the heterogeneous distributions inherent to human groups. Crucially, our mechanistic analysis suggests that this diversity collapse is not merely a behavioral anomaly but more likely a structural consequence of the low-rank bias inherent in neural network optimization. Therefore, our findings expose the limitations of current post-training paradigms and call for a shift toward alignment objectives that preserve cross-cultural pluralism. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-02 in cs.CL). Categorie arXiv: Social and Information Networks (cs.SI) ; Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2609.00565 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-17T06:33:38.885Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).