Automatic Context Pattern Generation for Entity Set Expansion

Entity Set Expansion (ESE) is a valuable task that aims to find entities of the target semantic class described by given seed entities. Various Natural Language Processing (NLP) and Information Retrieval (IR) downstream applications have benefited from ESE due its ability discover knowledge. Although existing corpus-based methods achieved great progress, they still rely on corpora with high-quality entity information annotated, because most them need obtain context patterns through position in sentence. Therefore, quality their annotation has become bottleneck limits performance such methods. To overcome this dilemma make models free dependence annotation, our work explore new paradigm, namely corpus-independent ESE. Specifically, we devise pattern generation module utilizes autoregressive language (e.g., GPT-2) automatically generate for In addition, propose GAPA, novel framework leverages aforementioned G ener xmlns:xlink="http://www.w3.org/1999/xlink">A ted xmlns:xlink="http://www.w3.org/1999/xlink">PA tterns expand Extensive experiments detailed analyses three widely used datasets demonstrate effectiveness method. All codes are available at https://github.com/geekjuruo/GAPA .