|
SensePOLAR: Word sense aware interpretability for pre-trained contextual word embeddings
Engler, Jan
;
Sikdar, Sandipan
;
Lutz, Marlene
;
Strohmaier, Markus
|
DOI:
|
https://doi.org/10.48550/ARXIV.2301.04704
|
|
URL:
|
https://aclanthology.org/2022.findings-emnlp.338
|
|
Document Type:
|
Conference or workshop publication
|
|
Year of publication:
|
2023
|
|
Book title:
|
Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing : EMNLP 2022
|
|
Page range:
|
4607-4619
|
|
Conference title:
|
EMNLP 2022
|
|
Location of the conference venue:
|
Abu Dhabi, United Arab Emirates
|
|
Date of the conference:
|
07.-11.12.2022
|
|
Publisher:
|
Goldberg, Yoav
;
Kozareva, Zornitsa
;
Zhang, Yue
|
|
Place of publication:
|
Stroudsburg, PA
|
|
Publishing house:
|
ACL
|
|
Related URLs:
|
|
|
Publication language:
|
English
|
|
Institution:
|
Business School > Data Science in the Economic and Social Sciences (Strohmaier, 2022-)
|
|
Subject:
|
400 Language, linguistics
|
|
Abstract:
|
Adding interpretability to word embeddings represents an area of active research in text representation. Recent work has explored thepotential of embedding words via so-called polar dimensions (e.g. good vs. bad, correct vs. wrong). Examples of such recent approaches include SemAxis, POLAR, FrameAxis, and BiImp. Although these approaches provide interpretable dimensions for words, they have not been designed to deal with polysemy, i.e. they can not easily distinguish between different senses of words. To address this limitation, we present SensePOLAR, an extension of the original POLAR framework that enables word-sense aware interpretability for pre-trained contextual word embeddings. The resulting interpretable word embeddings achieve a level of performance that is comparable to original contextual word embeddings across a variety of natural language processing tasks including the GLUE and SQuAD benchmarks. Our work removes a fundamental limitation of existing approaches by offering users sense aware interpretations for contextual word embeddings.
|
 | Dieser Eintrag ist Teil der Universitätsbibliographie. |
 Search Authors in
BASE:
Engler, Jan
;
Sikdar, Sandipan
;
Lutz, Marlene
;
Strohmaier, Markus
Google Scholar:
Engler, Jan
;
Sikdar, Sandipan
;
Lutz, Marlene
;
Strohmaier, Markus
ORCID:
Engler, Jan, Sikdar, Sandipan, Lutz, Marlene  ORCID: https://orcid.org/0000-0002-8265-2410 and Strohmaier, Markus ORCID: https://orcid.org/0000-0002-5485-5720
You have found an error? Please let us know about your desired correction here: E-Mail
Actions (login required)
 |
Show item |
|
|
