发表机构
Interdisciplinary School of Doctoral Studies; HLT Research Center; Faculty of Mathematics and Computer Science; Faculty of Foreign Languages and Literatures; University of Bucharest（跨学科博士研究院; HLT研究中心; 数学与计算机科学学院; 外国语与文学学院; 布加勒斯特大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究通过分析五种语言维基百科中三个罗马尼亚历史事件的叙事，揭示跨语言史学偏见和“引文孤立”现象，并提出“和平缔造者”流水线，利用对抗性提示实现自动化冲突调和。
AI 中文摘要
维基百科旨在提供统一、中立的历史记录，然而其各独立语言版本往往作为不同的认知共同体运作，围绕有争议的事件产生分歧性叙事。本文通过分析五种语言（罗马尼亚语、匈牙利语、俄语、土耳其语和英语）的维基百科文章，聚焦罗马尼亚历史上三个有争议的事件：波萨达战役（1330年）、苏联占领比萨拉比亚（1940年）和特尔戈维什泰夜袭（1462年），来研究跨语言史学偏见。我们使用人工标注者和大型语言模型（LLMs）对引文立场进行分类，并量化2005年至2024年间的叙事演变，识别出“引文孤立”现象。在波萨达战役的案例中，119条引文中仅有2条在不同语言版本间共享，罗马尼亚语版本表现出91%的亲民族偏见，而匈牙利语版本则相对平衡。纵向分析揭示这些叙事是易变的，并对当代地缘政治敏感，2024年俄语对比萨拉比亚的表述发生显著转变即为明证。最后，我们提出一个“和平缔造者”流水线来自动化冲突调和。我们证明，虽然标准提示会导致模型幻觉出共识，但“对抗性”提示——明确指示模型保留并归属分歧——能实现近乎完美的中立性得分。
英文摘要
Wikipedia aims to provide a unified, neutral record of history, yet its independent language editions often function as distinct epistemic communities, creating divergent narratives around contested events. This paper investigates cross-lingual historiographical bias by analyzing Wikipedia articles across five languages (Romanian, Hungarian, Russian, Turkish, and English) focusing on three contentious events in Romanian history: the Battle of Posada (1330), the Soviet occupation of Bessarabia (1940), and the Night Attack at Targoviste (1462). Using human annotators and Large Language Models (LLMs) to classify citation stance and quantify narrative evolution from 2005 to 2024, we identify a phenomenon of "citation isolation". In the case of the Battle of Posada, only 2 out of 119 citations were shared between language editions, with the Romanian edition exhibiting a 91% pro-national bias compared to the balanced Hungarian edition. Longitudinal analysis reveals that these narratives are volatile and responsive to contemporary geopolitics, evidenced by a significant shift in the Russian framing of Bessarabia in 2024. Finally, we propose a "Peace-Maker" pipeline to automate conflict reconciliation. We demonstrate that while standard prompting leads models to hallucinate consensus, "adversarial" prompting, which explicitly instructs the model to preserve and attribute disagreement, achieves near-perfect neutrality scores.
