SEE-Few: Seed, Expand and Entail for Few-shot Named Entity Recognition

Proceedings of the 29th International Conference on Computational Linguistics, pages 2540–2550 October 12–17, 2022. 2540 SEE-Few: Seed, Expand and Entail for Few-shot Named Entity Recognition Zeng Yang and Linhai Zhang and Deyu Zhou∗ School of Computer Science and Engineering, Key Laboratory of Computer Network and Information Integration, Ministry of Education, Southeast University, China {yangzeng, lzhang472, d.zhou}@seu.edu.cn Abstract Few-shot named entity recognition (NER) aims at identifying named entities based on only few labeled instances. Current few-shot NER meth- ods focus on leveraging existing datasets in the rich-resource domains which might fail in a training-from-scratch setting where no source- domain data is used. To tackle training-from- scratch setting, it is crucial to make full use of the annotation information (the boundaries and entity types). Therefore, in this paper, we propose a novel multi-task (Seed, Expand and Entail) learning framework, SEE-Few, for Few- shot NER without using source domain data. The seeding and expanding modules are respon- sible for providing as accurate candidate spans as possible for the entailing module. The entail- ing module reformulates span classification as a textual entailment task, leveraging both the con- textual clues and entity type information. All the three modules share the same text encoder and are jointly learned. Experimental results on four benchmark datasets under the training- from-scratch setting show that the proposed method outperformed state-of-the-art few-shot NER methods with a large margin. Our code is available at https://github.com/ unveiled-the-red-hat/SEE-Few. 1 Introduction Named entity recognition (NER), focusing on iden- tifying mention spans in text inputs and classifying them into the pre-defined entity categories, is a fundamental task in natural language processing and widely used in downstream tasks (Wang et al., 2019; Zhou et al., 2021; Peng et al., 2022). Super- vised NER has been intensively studied and yielded significant progress, especially with the aid of pre- trained language models (Devlin et al., 2019; Li et al., 2020; Mengge et al., 2020; Yu et al., 2020; Shen et al., 2021; Li et al., 2021a; Chen and Kong, 2021). However, supervised NER relies on plenty ∗Corresponding author. of training data, which is not suitable for some specific situations with few training data. Few-shot NER, aiming at recognizing entities based on few labeled instances, has attracted much attention in the research filed. Approaches for few- shot NER can be roughly divided into two cate- gories, span-based and sequence-labeling-based methods. Span-based approaches enumerate text spans in input texts and classify each span based on its corresponding template score (Cui et al., 2021). Sequence-labeling-based approaches treat NER as a sequence labeling problem which as- signs a tag for each token using the BIO or IO tag- ging scheme (Yang and Katiyar, 2020; Hou et al., 2020; Huang et al., 2021). Most of these span- based and sequence-labeling-based methods focus on leveraging existing datasets in the rich-resource domains to improve their performance in the low- resource domains. Unfortunately, the gap between the source domains and the target domains may hinder the performance of these methods (Pan and Yang, 2009; Cui et al., …

Content truncated to 500 words