Analyzing Text Representations by Measuring Task Alignment

Cesar Gonzalez-Gutierrez, Audi Primadhanty, Francesco Cazzaro, Ariadna Quattoni


Abstract
Textual representations based on pre-trained language models are key, especially in few-shot learning scenarios. What makes a representation good for text classification? Is it due to the geometric properties of the space or because it is well aligned with the task? We hypothesize the second claim. To test it, we develop a task alignment score based on hierarchical clustering that measures alignment at different levels of granularity. Our experiments on text classification validate our hypothesis by showing that task alignment can explain the classification performance of a given representation.
Anthology ID:
2023.acl-short.7
Volume:
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)
Month:
July
Year:
2023
Address:
Toronto, Canada
Editors:
Anna Rogers, Jordan Boyd-Graber, Naoaki Okazaki
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
70–81
Language:
URL:
https://aclanthology.org/2023.acl-short.7
DOI:
10.18653/v1/2023.acl-short.7
Bibkey:
Cite (ACL):
Cesar Gonzalez-Gutierrez, Audi Primadhanty, Francesco Cazzaro, and Ariadna Quattoni. 2023. Analyzing Text Representations by Measuring Task Alignment. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 70–81, Toronto, Canada. Association for Computational Linguistics.
Cite (Informal):
Analyzing Text Representations by Measuring Task Alignment (Gonzalez-Gutierrez et al., ACL 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.acl-short.7.pdf
Video:
 https://aclanthology.org/2023.acl-short.7.mp4