Named Entity Recognition (NER) in Few-Shot setting is imperative for entity tagging in low resource domains. Existing approaches only learn class-specific semantic features and intermediate representations from source domains. This affects generalizability to unseen target domains, resulting in suboptimal performances. To this end, we present CONTaiNER, a novel contrastive learning technique that optimizes the inter-token distribution distance for Few-Shot NER. Instead of optimizing class-specific attributes, CONTaiNER optimizes a generalized objective of differentiating between token categories based on their Gaussian-distributed embeddings. This effectively alleviates overfitting issues originating from training domains. Our experiments in several traditional test domains (OntoNotes, CoNLL'03, WNUT '17, GUM) and a new large scale Few-Shot NER dataset (Few-NERD) demonstrate that on average, CONTaiNER outperforms previous methods by 3%-13% absolute F1 points while showing consistent performance trends, even in challenging scenarios where previous approaches could not achieve appreciable performance.
翻译:在少光点设置中命名实体识别( NER) 是实体在低资源域标记的绝对必要条件。 现有方法只能从源域中学习类别特定的语义特征和中间表达方式。 这影响到普通到不可见的目标域, 导致亚优性表现。 为此, 我们展示了CONTAINER, 这是一种新颖的对比式学习技术, 优化了少光点 NER 的跨点分布距离 。 COTAINER 不仅没有优化特定类的属性, 还优化了基于其高频分布式嵌入的标志类别区别通用目标 。 这有效地缓解了来自培训域的问题。 我们在几个传统测试域的实验( Onto Notes, CoNLL'03, WNUT'17, GUM ) 和一个新的大规模“ 少光点 NER 数据集( Few- NERD) 显示, 平均来说, COTAINER 将先前的方法比3%- 13 % 绝对F1 点, 但也显示了一致的业绩趋势, 即使在以往方法无法取得显著业绩的有挑战性。