大语言模型中的比喻与文化知识：通过微调研究跨领域迁移
Figurative and Cultural Knowledge in LLMs: Investigating Cross-Domain Transfer through Fine-Tuning
- MBZUAI（穆罕默德·本·扎耶德人工智能大学）
- Carnegie Mellon University（卡内基梅隆大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
本研究通过在四款大语言模型上对阿拉伯文化类数据微调，发现诗歌微调可提升习语理解，文化微调则降低阿拉伯模型谚语理解，且跨领域迁移效果不佳。
AI中文摘要：
比喻语言深深植根于文化之中；流畅运用比喻语言不仅需要语言能力，还需要文化浸润。我们探究大语言模型（LLMs）能否学习这种关联：对文化数据进行微调能否提升比喻语言理解能力，反之亦然？我们在四个模型（ALLaM-7B、Fanar-1-9B、Qwen3-8B、Llama-3.1-8B）和六个阿拉伯数据集上开展了系统研究，这些数据集涵盖不同方言与地区的文化常识、谚语及诗歌。对诗歌进行微调可提升习语理解能力（+2.33%，p<0.05），而我们的ArabicMMLU对照实验未产生该增益，表明该增益源于比喻内容而非阿拉伯语适配，且显示出对非字面意义的敏感性可在比喻类型间迁移。相比之下，文化微调会降低两款阿拉伯中心模型的谚语理解准确率。此外，两个领域间的迁移与噪声无差异，阿拉伯模型在微调后常出现性能倒退，表明相关知识已处于先验饱和状态，而多语言模型则有更大的适应空间。错误分析进一步显示，微调会强化经验性文化知识，同时破坏基于历史的事实性知识。我们的研究结果表明，文化与比喻语言之间的关系虽在概念上合乎情理，但仅通过微调无法直接捕捉。
英文摘要：
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.
