Exploring Task Difficulty for Few-Shot Relation Extraction

Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 2605–2616 November 7–11, 2021. c⃝2021 Association for Computational Linguistics 2605 Exploring Task Difficulty for Few-Shot Relation Extraction Jiale Han1, Bo Cheng1 and Wei Lu2 1State Key Laboratory of Networking and Switching Technology, Beijing University of Posts and Telecommunications 2StatNLP Research Group, Singapore University of Technology and Design {hanjl,chengbo}@bupt.edu.cn, luwei@sutd.edu.sg Abstract Few-shot relation extraction (FSRE) focuses on recognizing novel relations by learning with merely a handful of annotated instances. Meta-learning has been widely adopted for such a task, which trains on randomly gen- erated few-shot tasks to learn generic data representations. Despite impressive results achieved, existing models still perform sub- optimally when handling hard FSRE tasks, where the relations are fine-grained and simi- lar to each other. We argue this is largely be- cause existing models do not distinguish hard tasks from easy ones in the learning process. In this paper, we introduce a novel approach based on contrastive learning that learns bet- ter representations by exploiting relation label information. We further design a method that allows the model to adaptively learn how to fo- cus on hard tasks. Experiments on two stan- dard datasets demonstrate the effectiveness of our method. 1 Introduction Relation extraction aims to detect the relation be- tween two entities contained in a sentence, which is the cornerstone of various natural language pro- cessing (NLP) applications, including knowledge base enrichment (Trisedya et al., 2019), biomed- ical knowledge discovery (Guo et al., 2020), and question answering (Han et al., 2020). Conven- tional neural methods (Miwa and Bansal, 2016; Tran et al., 2019) train a deep network through a large amount of labeled data with extensive rela- tions, so that the model can recognize these rela- tions during the test phase. Although impressive performance has been achieved, these methods are difficult to adapt to novel relations that have never been seen in the training process. In contrast, hu- mans can identify new relations with very few ex- amples. It is thus of great interest to enable the model to generalize to new relations with a handful of labeled instances. Figure 1: An example of easy few-shot task (top) and hard few-shot task (bottom). This is a 3-way-1-shot setup – each task involves three relations, and each re- lation has one supporting instance. Blue and red colors indicate head and tail entities respectively. For the easy task, the relations are very different, and it is easy to classify the query instance. However, due to the subtle differences among the relations in the hard tasks, it is challenging to correctly predict the true relation. Inspired by the success of few-shot learning in the computer vision (CV) community (Sung et al., 2018; Satorras and Estrach, 2018), Han et al. (2018) first introduce the task of few-shot relation extrac- tion (FSRE). FSRE requires models to be capable of handling classification of novel relations with scarce labeled instances. A popular framework for few-shot learning is meta-learning (Santoro et al., 2016; Vinyals …

Content truncated to 500 words