Fraunhofer SIT 在 CheckThat! 2023 中的参与：以 Check-Worthiness 分类为例，利用 Model Souping 应对分类不确定性
Fraunhofer SIT at CheckThat! 2023: Tackling Classification Uncertainty Using Model Souping on the Example of Check-Worthiness Classification
AI总结：
Fraunhofer SIT 团队在 CLEF-2023 CheckThat! Task 1B 中提出基于 Model Souping 的集成分类方法，用于政治辩论文本的值得核查性分类，在英语数据集上取得 0.878 的 F1 分数，排名第二。
摘要：
本文描述了 Fraunhofer SIT 团队在 CLEF-2023 CheckThat! 实验室 Task 1B（英语）中开发的第二名方法。给定一段来自政治辩论的文本片段，该任务的目标是确定其是否应被评估为值得核查。检测值得核查的陈述旨在通过优先考虑事实核查员应首先考虑的声明来促进人工事实核查工作。它也可以被视为事实核查系统的首要步骤。我们表现最佳的方法利用了以 Model Souping 为中心的集成分类方案。当应用于英语数据集时，我们提交的模型取得了 0.878 的总体 F1 分数，并在比赛中排名第二。
英文摘要：
This paper describes the second-placed approach developed by the Fraunhofer SIT team in the CLEF-2023 CheckThat! lab Task 1B for English. Given a text snippet from a political debate, the aim of this task is to determine whether it should be assessed for check-worthiness. Detecting check-worthy statements aims to facilitate manual fact-checking efforts by prioritizing the claims that fact-checkers should consider first. It can also be considered as primary step of a fact-checking system. Our best-performing method took advantage of an ensemble classification scheme centered on Model Souping. When applied to the English data set, our submitted model achieved an overall F1 score of 0.878 and was ranked as the second-best model in the competition.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
