Probing Cultural Awareness in LLMs: A Case Study of Cross-Culture Aesthetic Stylistics

Wang, Jiashuo; Yu, Fenggang; Wang, Jian; Leong, Chak Tou; Shen, Xiaoyu; Xu, Chunpu; Duan, Jiawen; Li, Wenjie; Hoorn, Johan F.

doi:10.24963/ijcai.2026/856

Probing Cultural Awareness in LLMs: A Case Study of Cross-Culture Aesthetic Stylistics

Jiashuo Wang, Fenggang Yu, Jian Wang, Chak Tou Leong, Xiaoyu Shen, Chunpu Xu, Jiawen Duan, Wenjie Li, Johan F. Hoorn

Download BibTeX Cite

Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence. Human-Centred AI. Pages 7703-7711.

https://doi.org/10.24963/ijcai.2026/856

Keywords

Human-Centred AI: Humans and AI

Abstract

Large Language Models (LLMs) are increasingly deployed in diverse cultural contexts, yet their ability to master aesthetic stylistics, i.e., the strategic use of language to evoke cultural resonance, remains underexplored. We curate C4Styli, a benchmark of highly stylized translated movie titles and advertising slogans from Hong Kong and the Chinese Mainland, to evaluate LLMs via the lens of behavioral recognition and productive competence. Extensive evaluations show that LLMs differ from humans in stylistic recognition, and this recognition ability varies across text domains. In addition, stylistic recognition and generation performance in LLMs are not consistently aligned. To further examine whether LLMs genuinely capture stylistic information in stylistic recognition, we conduct structural ablation with logistic regression probes. We find that, in the Hong Kong setting, stylistic recognition in LLMs relies primarily on surface-level linguistic information rather than stylistic structure. This suggests limited sensitivity to Hong Kong-specific stylistic structure. Our

code and data are available at https://github.com/wangjs9/C4STYLI.

Jiashuo Wang, Fenggang Yu, Jian Wang, Chak Tou Leong, Xiaoyu Shen, Chunpu Xu, Jiawen Duan, Wenjie Li, and Johan F. Hoorn. 2026. Probing Cultural Awareness in LLMs: A Case Study of Cross-Culture Aesthetic Stylistics. In Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26. IJCAI, 7703–7711. https://doi.org/10.24963/ijcai.2026/856

@inproceedings{ijcai2026p856,
  title     = {Probing Cultural Awareness in LLMs: A Case Study of Cross-Culture Aesthetic Stylistics},
  author    = {Wang, Jiashuo and Yu, Fenggang and Wang, Jian and Leong, Chak Tou and Shen, Xiaoyu and Xu, Chunpu and Duan, Jiawen and Li, Wenjie and Hoorn, Johan F.},
  booktitle = {Proceedings of the Thirty-Fifth International Joint Conference on
               Artificial Intelligence, {IJCAI-26}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {Diego Calvanese},
  pages     = {7703--7711},
  year      = {2026},
  month     = {8},
  note      = {Human-Centred AI},
  doi       = {10.24963/ijcai.2026/856},
  url       = {https://doi.org/10.24963/ijcai.2026/856},
}

TY  - CPAPER
TI  - Probing Cultural Awareness in LLMs: A Case Study of Cross-Culture Aesthetic Stylistics
AU  - Wang, Jiashuo
AU  - Yu, Fenggang
AU  - Wang, Jian
AU  - Leong, Chak Tou
AU  - Shen, Xiaoyu
AU  - Xu, Chunpu
AU  - Duan, Jiawen
AU  - Li, Wenjie
AU  - Hoorn, Johan F.
T2  - Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2026
SP  - 7703
EP  - 7711
DO  - 10.24963/ijcai.2026/856
UR  - https://doi.org/10.24963/ijcai.2026/856
ER  -