AKCIT-FN at CheckThat! 2025：切换微调的 SLM 与 LLM 提示以进行多语言主张规范化
AKCIT-FN at CheckThat! 2025: Switching Fine-Tuned SLMs and LLM Prompting for Multilingual Claim Normalization
AI总结：
本文提出结合微调小型语言模型和大语言模型提示的方法，在 CLEF-2025 CheckThat! 多语言主张规范化任务的二十种语言中十五种获得前三名，验证了零样本策略的有效性。
摘要：
主张规范化是将非正式社交媒体帖子转化为简洁、独立陈述的过程，是自动化事实核查流程中的关键步骤。本文详细介绍了我们提交至 CLEF-2025 CheckThat! Task 2 的系统，该任务要求系统在二十种语言中执行主张规范化，分为十三个有监督（高资源）和七个零样本（无训练数据）赛道。我们的方法在有监督语言中利用微调的小型语言模型（SLM），在零样本场景中利用大语言模型（LLM）提示，在二十种语言中的十五种中获得了领奖台名次（前三名）。值得注意的是，这包括在八种语言中获得第二名，其中五种属于指定的七种零样本语言，突显了我们基于 LLM 的零样本策略的有效性。对于葡萄牙语（我们的初始开发语言），系统取得了 0.5290 的平均 METEOR 分数，排名第三。所有实现成果（包括推理、训练、评估脚本和提示配置）均公开于 https://github.com/ju-resplande/checkthat2025_normalization。
英文摘要：
Claim normalization, the transformation of informal social media posts into concise, self-contained statements, is a crucial step in automated fact-checking pipelines. This paper details our submission to the CLEF-2025 CheckThat! Task~2, which challenges systems to perform claim normalization across twenty languages, divided into thirteen supervised (high-resource) and seven zero-shot (no training data) tracks. Our approach, leveraging fine-tuned Small Language Models (SLMs) for supervised languages and Large Language Model (LLM) prompting for zero-shot scenarios, achieved podium positions (top three) in fifteen of the twenty languages. Notably, this included second-place rankings in eight languages, five of which were among the seven designated zero-shot languages, underscoring the effectiveness of our LLM-based zero-shot strategy. For Portuguese, our initial development language, our system achieved an average METEOR score of 0.5290, ranking third. All implementation artifacts, including inference, training, evaluation scripts, and prompt configurations, are publicly available at https://github.com/ju-resplande/checkthat2025_normalization.
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。
