研究人员推出了PubMedCausal，一个用于生物医学文本因果关系抽取的新语料库。该数据集源自PubMed摘要，提供了3,945个因果关系行和6,491个因果对的跨度级标注，能够对模型能力进行详细评估。基准测试表明，虽然像PubMedBERT这样的生物医学编码器在因果检测方面表现强劲，但像DeepSeek-R1-32B这样的生成模型在少样本提示下进行跨度级抽取时也能取得有竞争力的结果。 AI
影响 该语料库将能够更精确地评估AI模型在理解生物医学文献中复杂因果关系方面的能力。
排序理由 该集群描述了一篇介绍用于生物医学领域特定NLP任务的新型标注语料库的学术论文。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2605.28363v1 Announce Type: new Abstract: Causal relation extraction (CRE) is central to biomedical text mining, but current resources often conflate causal relations with broader associations, restrict annotation to sentence-level examples, or focus mainly on explicit caus…
Causal relation extraction (CRE) is central to biomedical text mining, but current resources often conflate causal relations with broader associations, restrict annotation to sentence-level examples, or focus mainly on explicit causal cues. This limits their usefulness for evalua…