轻量级模型在少量标注数据下实现高精度生物医学实体识别
ReProCon: Scalable and Resource-Efficient Few-Shot Biomedical Named Entity Recognition
- 用多个原型+对比学习捕捉实体语义差异，缓解类别不平衡
- 仅需30%标签数据仍保持稳定，50类时F1仅降7.8%
- 适合资源受限的生物医学文本分析场景
生物医学命名实体识别面临数据稀缺和标签分布不均的问题，尤其在细粒度实体类型上。我们提出ReProCon框架，结合多原型建模、余弦对比学习与Reptile元学习，有效应对这些挑战。通过为每类实体设置多个原型，模型能捕捉同义词和上下文差异；余弦对比目标强化类别间分离。采用轻量级fastText + BiLSTM编码器，内存占用低，宏F1接近基于BERT的基线（约达其99%）。在标签预算30%时性能稳定，类别从19增至50时F1仅下降7.8%，显著优于SpanProto和CONTaiNER等基线（后者退化10–32%）。消融实验验证了多原型与对比学习对缓解类别不平衡的关键作用。尽管存在标签歧义，该模型在资源受限条件下表现优异，适用于生物医学应用。
原文摘要 · Abstract (English)
Named Entity Recognition (NER) in biomedical domains faces challenges due to data scarcity and imbalanced label distributions, especially with fine-grained entity types. We propose ReProCon, a novel few-shot NER framework that combines multi-prototype modeling, cosine-contrastive learning, and Reptile meta-learning to tackle these issues. By representing each category with multiple prototypes, ReProCon captures semantic variability, such as synonyms and contextual differences, while a cosine-contrastive objective ensures strong interclass separation. Reptile meta-updates enable quick adaptation with little data. Using a lightweight fastText + BiLSTM encoder with much lower memory usage, ReProCon achieves a macro-$F_1$ score close to BERT-based baselines (around 99 percent of BERT performance). The model remains stable with a label budget of 30 percent and only drops 7.8 percent in $F_1$ when expanding from 19 to 50 categories, outperforming baselines such as SpanProto and CONTaiNER, which see 10 to 32 percent degradation in Few-NERD. Ablation studies highlight the importance of multi-prototype modeling and contrastive learning in managing class imbalance. Despite difficulties with label ambiguity, ReProCon demonstrates state-of-the-art performance in resource-limited settings, making it suitable for biomedical applications.
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。