Culturally-Nuanced Story Generation for Reasoning in Low-Resource Languages: The Case of Javanese and Sundanese
Pranida, Salsabila Zahirah Genadi, Rifo Ahmad Koto, Fajri
Pranida, Salsabila Zahirah
Genadi, Rifo Ahmad
Koto, Fajri
Supervisor
Department
Natural Language Processing
Embargo End Date
Type
Conference proceeding
Date
2025
License
Language
English
Collections
Research Projects
Organizational Units
Journal Issue
Abstract
Culturally grounded commonsense reasoning is underexplored in low-resource languages due to scarce data and costly native annotation. We test whether large language models (LLMs) can generate culturally nuanced narratives for such settings. Focusing on Javanese and Sundanese, we compare three data creation strategies: (1) LLM-assisted stories prompted with cultural cues, (2) machine translation from Indonesian benchmarks, and (3) native-written stories. Human evaluation finds LLM stories match natives on cultural fidelity but lag in coherence and correctness. We fine-tune models on each dataset and evaluate on a human-authored test set for classification and generation. LLM-generated data yields higher downstream performance than machine-translated and Indonesian human-authored training data. We release a high-quality benchmark of culturally grounded commonsense stories in Javanese and Sundanese to support future work.
Citation
S. Z. Pranida, R. A. Genadi, and F. Koto, “Culturally-Nuanced Story Generation for Reasoning in Low-Resource Languages: The Case of Javanese and Sundanese,” Proceedings of the 5th Workshop on Multilingual Representation Learning (MRL 2025), pp. 369–384, 2025, doi: 10.18653/V1/2025.MRL-MAIN.25.
Source
Proceedings of the 5th Workshop on Multilingual Representation Learning (MRL 2025)
Conference
5th Workshop on Multilingual Representation Learning (MRL 2025)
Keywords
Multimodal Retrieval, Language Models, Cross-Modal Alignment, Visual-Textual Embeddings, Contrastive Learning, Zero-Shot Retrieval, Unified Retrieval Benchmark, Scalable Embedding Indexing
Subjects
Source
5th Workshop on Multilingual Representation Learning (MRL 2025)
Publisher
Association for Computational Linguistics