Yuan Yao

TL;DR: This paper presented a few-shot relation classification dataset (FewRel) consisting of 70, 000 sentences on 100 relations derived from Wikipedia and annotated by crowdworkers. And they adapted the most recent state-of-the-art fewshot learning methods for relation classification and conduct a thorough evaluation of these methods.

TL;DR: This paper presented a few-shot relation classification dataset, consisting of 70, 000 sentences on 100 relations derived from Wikipedia and annotated by crowdworkers, where the relation of each sentence is first recognized by distant supervision methods, and then filtered by crowd workers.

TL;DR: DocRED as mentioned in this paper is a document-level relation extraction dataset constructed from Wikipedia and Wikidata with three features: (1) DocRED annotates both named entities and relations, and is the largest human-annotated dataset for documentlevel RE from plain text; (2) DocRed requires reading multiple sentences in a document to extract entities and infer their relations by synthesizing all information of the document; and (3) along with the humanannotated data, which enables DocRED to be adopted for both supervised and weakly supervised scenarios.

TL;DR: Empirical results show that DocRED is challenging for existing RE methods, which indicates that document-level RE remains an open problem and requires further efforts.

TL;DR: An online system to meet real-time extraction without any training and deploying and can extract facts in various scenarios as well as aligning the extracted facts to Wikidata, which may benefit various downstream knowledge-driven applications.