Figurative and Cultural Knowledge in LLMs: Investigating Cross-Domain Transfer through Fine-Tuning

• URL: http://arxiv.org/abs/2608.18361v1
• Date: Tue, 18 Aug 2026 22:25:17 GMT
• Title: Figurative and Cultural Knowledge in LLMs: Investigating Cross-Domain Transfer through Fine-Tuning
• Abstract summary: Figurative language is deeply culturally embedded; fluent use requires cultural immersion.<n>We ask whether fine-tuning on cultural data improve figurative language understanding, and vice versa.<n>Our findings suggest that the relationship between culture and figurative language, though conceptually natural, is not straightforwardly captured through fine-tuning alone.
• Score: 11.498685324357384
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Figurative language is deeply culturally embedded; fluent use requires not just linguistic competence but cultural immersion. We ask whether LLMs can learn this link: does fine-tuning on cultural data improve figurative language understanding, and vice versa? We conduct a systematic study across four models (ALLaM-7B, Fanar-1-9B, Qwen3-8B, Llama-3.1-8B) and six Arabic datasets spanning cultural commonsense, proverbs, and poetry across diverse dialects and regions. Fine-tuning on poetry improves idiom comprehension (+2.33%, p<0.05), a gain our ArabicMMLU control does not reproduce, indicating that it stems from figurative content rather than Arabic language adaptation and pointing to a sensitivity to non-literal meaning that transfers across figurative types. Cultural fine-tuning, by contrast, lowers proverb-interpretation accuracy in both Arabic-centric models. Transfer between the two domains is otherwise indistinguishable from noise, with Arabic models frequently regressing after fine-tuning, suggesting prior saturation of relevant knowledge, while multilingual models show greater adaptation headroom. Error analysis further reveals that fine-tuning reinforces experiential cultural knowledge while destabilizing historically grounded factual knowledge. Our findings suggest that the relationship between culture and figurative language, though conceptually natural, is not straightforwardly captured through fine-tuning alone.

Related papers

• Same Lesson, Different Story: Cross-Lingual Reconstruction of Cultural Narratives in Large Language Models [0.0]
  This study introduces a multilingual evaluation narrative framework that integrates a cross-linguistic collection of 414 proverbs spanning 15 languages.<n>The analysis assesses whether models preserve meaning across languages, how cross-lingual conditioning influences narrative realization, and whether different model families converge on similar interpretations.
  arXiv  Detail & Related papers  (2026-06-23T14:10:34Z)
• Where Culture Fades: Revealing the Cultural Gap in Text-to-Image Generation [43.352493955825736]
  We show that current T2I models often produce culturally neutral or English-biased results under multilingual prompts.<n>We propose a probing method that localizes culture-sensitive signals to a small set of neurons in a few fixed layers.
  arXiv  Detail & Related papers  (2025-11-21T14:40:50Z)
• Do You Know About My Nation? Investigating Multilingual Language Models' Cultural Literacy Through Factual Knowledge [68.6805229085352]
  Most multilingual question-answering benchmarks do not factor in regional diversity in the information they capture.<n>XNationQA encompasses a total of 49,280 questions on the geography, culture, and history of nine countries, presented in seven languages.<n>We benchmark eight standard multilingual LLMs on XNationQA and evaluate them using two novel transference metrics.
  arXiv  Detail & Related papers  (2025-11-01T18:41:34Z)
• 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)
• Jawaher: A Multidialectal Dataset of Arabic Proverbs for LLM Benchmarking [12.078532717928185]
  Large language models (LLMs) continue to exhibit biases toward Western, Anglo-centric, or American cultures.<n>We introduce Jawaher, a benchmark designed to assess LLMs' capacity to comprehend and interpret Arabic proverbs.<n>We find that while LLMs can generate idiomatically accurate translations, they struggle with producing culturally nuanced and contextually relevant explanations.
  arXiv  Detail & Related papers  (2025-02-28T22:28:00Z)
• On The Origin of Cultural Biases in Language Models: From Pre-training Data to Linguistic Phenomena [10.263201685476492]
  This paper aims to uncover the origins of entity-related cultural biases in Language Models (LMs)<n>We introduce CAMeL-2, a parallel Arabic-English benchmark of 58,086 entities associated with Arab and Western cultures and 367 masked natural contexts for entities.<n>Our evaluations using CAMeL-2 reveal reduced performance gaps between cultures by LMs when tested in English compared to Arabic.
  arXiv  Detail & Related papers  (2025-01-08T18:15:47Z)
• Bilingual Adaptation of Monolingual Foundation Models [48.859227944759986]
  We present an efficient method for adapting a monolingual Large Language Model (LLM) to another language. Our two-stage approach begins with expanding the vocabulary and training only the embeddings matrix. By continually pre-training on a mix of Arabic and English corpora, the model retains its proficiency in English while acquiring capabilities in Arabic.
  arXiv  Detail & Related papers  (2024-07-13T21:09:38Z)
• See It from My Perspective: How Language Affects Cultural Bias in Image Understanding [60.70852566256668]
  Vision-language models (VLMs) can respond to queries about images in many languages.<n>We characterize the Western bias of VLMs in image understanding and investigate the role that language plays in this disparity.
  arXiv  Detail & Related papers  (2024-06-17T15:49:51Z)
• Not All Countries Celebrate Thanksgiving: On the Cultural Dominance in Large Language Models [89.94270049334479]
  This paper identifies a cultural dominance issue within large language models (LLMs) LLMs often provide inappropriate English-culture-related answers that are not relevant to the expected culture when users ask in non-English languages.
  arXiv  Detail & Related papers  (2023-10-19T05:38:23Z)
• Multi-lingual and Multi-cultural Figurative Language Understanding [69.47641938200817]
  Figurative language permeates human communication, but is relatively understudied in NLP. We create a dataset for seven diverse languages associated with a variety of cultures: Hindi, Indonesian, Javanese, Kannada, Sundanese, Swahili and Yoruba. Our dataset reveals that each language relies on cultural and regional concepts for figurative expressions, with the highest overlap between languages originating from the same region. All languages exhibit a significant deficiency compared to English, with variations in performance reflecting the availability of pre-training and fine-tuning data.
  arXiv  Detail & Related papers  (2023-05-25T15:30:31Z)
• Having Beer after Prayer? Measuring Cultural Bias in Large Language Models [25.722262209465846]
  We show that multilingual and Arabic monolingual LMs exhibit bias towards entities associated with Western culture. We introduce CAMeL, a novel resource of 628 naturally-occurring prompts and 20,368 entities spanning eight types that contrast Arab and Western cultures. Using CAMeL, we examine the cross-cultural performance in Arabic of 16 different LMs on tasks such as story generation, NER, and sentiment analysis.
  arXiv  Detail & Related papers  (2023-05-23T18:27:51Z)

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.