发表机构
University of Ottawa（渥太华大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文介绍渥太华大学团队通过对gemini-2.5-pro和claude-sonnet-4-5进行提示工程，利用跨语言迁移学习完成古典拉丁语NER任务，在EvaLatin 2026的两个NER子任务中均获第一。
AI 中文摘要
随着古典拉丁语文本数字化资源的增加以及大语言模型（LLMs）的现代突破，本文作者通过参与EvaLatin 2026，为古代语言研究作出贡献。本文介绍了渥太华大学（Team uOttawa）在命名实体识别（NER）共享任务中的系统描述与结果。该任务分为两个子任务：包含11个类别的粗粒度NER，以及包含28个类别的细粒度NER，每个子任务均在严格和模糊两种评估机制下进行。通过对商业LLMs gemini-2.5-pro和claude-sonnet-4-5进行提示工程，本文表明，代表性不足的古代拉丁语可以利用跨语言迁移学习，借助更广泛LLM开发社区的进步来实现发展。总体而言，本文讨论的方法展现出非常出色的结果，在两个NER子任务中均排名第一，且在所有提交作品的所有评估指标和机制中均取得了最佳分数。
英文摘要
With the increase in digitized resources of Classical Latin texts and modern breakthroughs of Large Language Models (LLMs), I contribute to ancient language research by participating in EvaLatin 2026. This paper describes Team uOttawa's system description and results for the Named Entity Recognition (NER) shared task. The task is divided into two subtasks: coarse-grained NER with 11 classes and fine-grained NER with 28 classes, each evaluated under strict and fuzzy regimes. Through prompt engineering of commercial LLMs gemini-2.5-pro and claude-sonnet-4-5, I show that the underrepresented ancient Latin language can take advantage of cross-lingual transfer learning by using advancements made by the wider LLM development community. Overall, the methods discussed in this report demonstrate very strong results, placing first in both NER subtasks and achieving the best scores across all evaluation metrics and regimes among all submissions.
Journal refEvaLatin (LT4HALA@LREC), ELRA, May 2026, Palma De Majorque, Spain
