用微调Transformer和大模型提示词，提升多语言主观性检测效果
XplaiNLP at CheckThat! 2025: Multilingual Subjectivity Detection with Finetuned Transformers and Prompt-Based Inference with Large Language Models
- 微调EuroBERT等模型，结合翻译数据提升性能
- 意大利语任务达F1 0.8104，超越基线0.6941
- 适合多语言文本分析与跨语言主观性研究者
本报告介绍XplaiNLP在CheckThat! 2025多语言主观性检测共享任务中的参赛方案。我们评估两种方法：(1) 在单语及机器翻译训练数据上对EuroBERT、XLM-RoBERTa、German-BERT进行监督微调；(2) 使用大模型零样本提示：o3-mini用于标注（规则标注），gpt-4.1-mini用于DoubleDown（对比重写）和Perspective（比较推理）。在意大利语单语子任务中，标注方法取得第一，F1达0.8104，优于基线0.6941。罗马尼亚零样本设置下，微调的XLM-RoBERTa模型获得F1 0.7917，排名第三，超过基线0.6461。该模型在多语言任务中表现稳定，希腊语也优于基线。德语方面，基于相关语言翻译数据微调的German-BERT表现良好。但乌克兰和波兰的零样本表现略低于基线，体现低资源跨语言泛化的挑战。
原文摘要 · Abstract (English)
This notebook reports the XplaiNLP submission to the CheckThat! 2025 shared task on multilingual subjectivity detection. We evaluate two approaches: (1) supervised fine-tuning of transformer encoders, EuroBERT, XLM-RoBERTa, and German-BERT, on monolingual and machine-translated training data; and (2) zero-shot prompting using two LLMs: o3-mini for Annotation (rule-based labelling) and gpt-4.1-mini for DoubleDown (contrastive rewriting) and Perspective (comparative reasoning). The Annotation Approach achieves 1st place in the Italian monolingual subtask with an F_1 score of 0.8104, outperforming the baseline of 0.6941. In the Romanian zero-shot setting, the fine-tuned XLM-RoBERTa model obtains an F_1 score of 0.7917, ranking 3rd and exceeding the baseline of 0.6461. The same model also performs reliably in the multilingual task and improves over the baseline in Greek. For German, a German-BERT model fine-tuned on translated training data from typologically related languages yields competitive performance over the baseline. In contrast, performance in the Ukrainian and Polish zero-shot settings falls slightly below the respective baselines, reflecting the challenge of generalization in low-resource cross-lingual scenarios.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。