Proxy-based Zero-Shot Entity Linking by Effective Candidate Retrieval

Maciej Wiatrak, Eirini Arvaniti, Angus Brayne, Jonas Vetterle, Aaron Sim


Abstract
A recent advancement in the domain of biomedical Entity Linking is the development of powerful two-stage algorithms – an initial candidate retrieval stage that generates a shortlist of entities for each mention, followed by a candidate ranking stage. However, the effectiveness of both stages are inextricably dependent on computationally expensive components. Specifically, in candidate retrieval via dense representation retrieval it is important to have hard negative samples, which require repeated forward passes and nearest neighbour searches across the entire entity label set throughout training. In this work, we show that pairing a proxy-based metric learning loss with an adversarial regularizer provides an efficient alternative to hard negative sampling in the candidate retrieval stage. In particular, we show competitive performance on the recall@1 metric, thereby providing the option to leave out the expensive candidate ranking step. Finally, we demonstrate how the model can be used in a zero-shot setting to discover out of knowledge base biomedical entities.
Anthology ID:
2022.louhi-1.11
Volume:
Proceedings of the 13th International Workshop on Health Text Mining and Information Analysis (LOUHI)
Month:
December
Year:
2022
Address:
Abu Dhabi, United Arab Emirates (Hybrid)
Editors:
Alberto Lavelli, Eben Holderness, Antonio Jimeno Yepes, Anne-Lyse Minard, James Pustejovsky, Fabio Rinaldi
Venue:
Louhi
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
87–99
Language:
URL:
https://aclanthology.org/2022.louhi-1.11
DOI:
10.18653/v1/2022.louhi-1.11
Bibkey:
Cite (ACL):
Maciej Wiatrak, Eirini Arvaniti, Angus Brayne, Jonas Vetterle, and Aaron Sim. 2022. Proxy-based Zero-Shot Entity Linking by Effective Candidate Retrieval. In Proceedings of the 13th International Workshop on Health Text Mining and Information Analysis (LOUHI), pages 87–99, Abu Dhabi, United Arab Emirates (Hybrid). Association for Computational Linguistics.
Cite (Informal):
Proxy-based Zero-Shot Entity Linking by Effective Candidate Retrieval (Wiatrak et al., Louhi 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.louhi-1.11.pdf
Video:
 https://aclanthology.org/2022.louhi-1.11.mp4