Abstract
In morphologically rich languages, words consist of morphemes containing deeper information in morphology, and thus such languages may necessitate the use of morpheme-level representations as well as word representations. This study introduces a neural multilingual end-to-end coreference resolution system by incorporating morphological information in transformer-based word embeddings on the baseline model. This proposed model participated in the Sixth Workshop on Computational Models of Reference, Anaphora and Coreference (CRAC 2023). Including morphological information explicitly into the coreference resolution improves the performance, especially in morphologically rich languages (e.g., Catalan, Hungarian, and Turkish). The introduced model outperforms the baseline system by 2.57 percentage points on average by obtaining 59.53% CoNLL F-score.
Original language | English |
---|---|
Title of host publication | EMNLP 2023 - Proceedings of the CRAC 2023 Shared Task on Multilingual Coreference Resolution at 6th Workshop on Computational Models of Reference, Anaphora and Coreference, CRAC 2023 |
Editors | Zabokrtsky Zabokrtsky, Maciej Ogrodniczuk |
Publisher | Association for Computational Linguistics (ACL) |
Pages | 34-40 |
Number of pages | 7 |
ISBN (Electronic) | 9781955917025 |
Publication status | Published - 2023 |
Event | CRAC Shared Task on Multilingual Coreference Resolution at 6th Workshop on Computational Models of Reference, Anaphora and Coreference, CRAC 2023 - Singapore, Singapore Duration: 7 Dec 2023 → … |
Publication series
Name | EMNLP 2023 - Proceedings of the CRAC 2023 Shared Task on Multilingual Coreference Resolution at 6th Workshop on Computational Models of Reference, Anaphora and Coreference, CRAC 2023 |
---|
Conference
Conference | CRAC Shared Task on Multilingual Coreference Resolution at 6th Workshop on Computational Models of Reference, Anaphora and Coreference, CRAC 2023 |
---|---|
Country/Territory | Singapore |
City | Singapore |
Period | 7/12/23 → … |
Bibliographical note
Publisher Copyright:© CRAC 2023. All Rights Reserved.
Funding
This work is funded by the Scientific and Technological Research Council of Turkey (TUBITAK) with a TUBITAK 2515 (European Cooperation in Science and Technology - COST) project Grant No. 123E079. Computing resources used in this work were provided by the National Center for High Performance Computing of Turkey (UHeM) under grant number 4015042023 and also by İTÜ Artificial Intelligence and Data Science Application and Research Center. This work is funded by the Scientific and Technological Research Council of Turkey (TUBITAK) with a TUBITAK 2515 (European Cooperation in Science and Technology - COST) project Grant No. 123E079. Computing resources used in this work were provided by the National Center for High Performance Computing of Turkey (UHeM) under grant number 4015042023 and also by ˙TÜ Artificial Intelligence and Data Science Application and Research Center.
Funders | Funder number |
---|---|
İTÜ Artificial Intelligence and Data Science Application and Research Center | |
National Center for High Performance Computing of Turkey | |
TUBITAK 2515 | |
Ulusal Yüksek Başarımlı Hesaplama Merkezi, Istanbul Teknik Üniversitesi | 4015042023 |
European Cooperation in Science and Technology | 123E079 |
Türkiye Bilimsel ve Teknolojik Araştırma Kurumu | |
˙TÜ Artificial Intelligence and Data Science Application and Research Center |