2 papers were accepted at the Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL), a top NLP conference to be held on July 10th-15th, 2022.

・EASE: Entity-Aware Contrastive Learning of Sentence Embedding: We propose a novel method for learning sentence embeddings via contrastive learning between sentences and their related entities.

・Global Entity Disambiguation with BERT: We propose a global entity disambiguation model based on BERT.

NAACL 2022 (full paper)

EASE: Entity-Aware Contrastive Learning of Sentence Embedding

Sosuke Nishikawa (Studio Ousia/The University of Tokyo), Ryokan Ri (Studio Ousia/The University of Tokyo), Ikuya Yamada, Yoshimasa Tsuruoka (The University of Tokyo), Isao Echizen (National Institute of Informatics)

NAACL 2022 (short paper)

Global Entity Disambiguation with BERT

Ikuya Yamada, Koki Washio (Megagon Labs) , Hiroyuki Shindo (NAIST, RIKEN), Yuji Matsumoto (RIKEN)

https://arxiv.org/abs/1909.00426