Few-shot Named Entity Recognition with Cloze Questions
This addresses the challenge of data scarcity in NER for low-resource languages and domains, though it is incremental as it adapts an existing approach.
The paper tackles the problem of limited annotated data for Named Entity Recognition (NER) by proposing a few-shot learning method using cloze questions, achieving better performance than standard fine-tuning and competitive results on benchmark datasets without manual annotations.
Despite the huge and continuous advances in computational linguistics, the lack of annotated data for Named Entity Recognition (NER) is still a challenging issue, especially in low-resource languages and when domain knowledge is required for high-quality annotations. Recent findings in NLP show the effectiveness of cloze-style questions in enabling language models to leverage the knowledge they acquired during the pre-training phase. In our work, we propose a simple and intuitive adaptation of Pattern-Exploiting Training (PET), a recent approach which combines the cloze-questions mechanism and fine-tuning for few-shot learning: the key idea is to rephrase the NER task with patterns. Our approach achieves considerably better performance than standard fine-tuning and comparable or improved results with respect to other few-shot baselines without relying on manually annotated data or distant supervision on three benchmark datasets: NCBI-disease, BC2GM and a private Italian biomedical corpus.