Skip to main content

Improving Downstream Task Performance by Treating Numbers as Entities

Publication ,  Conference
Sundararaman, D; Subramanian, V; Wang, G; Xu, L; Carin, L
Published in: International Conference on Information and Knowledge Management, Proceedings
October 17, 2022

Numbers are essential components of text, like any other word tokens, from which natural language processing (NLP) models are built and deployed. Though numbers are typically not accounted for distinctly in most NLP tasks, there is still an underlying amount of numeracy already exhibited by NLP models. For instance, in named entity recognition (NER), numbers are not treated as an entity with distinct tags. In this work, we attempt to tap the potential of state-of-the-art language models and transfer their ability to boost performance in related downstream tasks dealing with numbers. Our proposed classification of numbers into entities helps NLP models perform well on several tasks, including a handcrafted Fill-In-The-Blank (FITB) task and on question answering, using joint embeddings, outperforming the BERT and RoBERTa baseline classification.

Duke Scholars

Published In

International Conference on Information and Knowledge Management, Proceedings

DOI

Publication Date

October 17, 2022

Start / End Page

4535 / 4539
 

Citation

APA
Chicago
ICMJE
MLA
NLM
Sundararaman, D., Subramanian, V., Wang, G., Xu, L., & Carin, L. (2022). Improving Downstream Task Performance by Treating Numbers as Entities. In International Conference on Information and Knowledge Management, Proceedings (pp. 4535–4539). https://doi.org/10.1145/3511808.3557614
Sundararaman, D., V. Subramanian, G. Wang, L. Xu, and L. Carin. “Improving Downstream Task Performance by Treating Numbers as Entities.” In International Conference on Information and Knowledge Management, Proceedings, 4535–39, 2022. https://doi.org/10.1145/3511808.3557614.
Sundararaman D, Subramanian V, Wang G, Xu L, Carin L. Improving Downstream Task Performance by Treating Numbers as Entities. In: International Conference on Information and Knowledge Management, Proceedings. 2022. p. 4535–9.
Sundararaman, D., et al. “Improving Downstream Task Performance by Treating Numbers as Entities.” International Conference on Information and Knowledge Management, Proceedings, 2022, pp. 4535–39. Scopus, doi:10.1145/3511808.3557614.
Sundararaman D, Subramanian V, Wang G, Xu L, Carin L. Improving Downstream Task Performance by Treating Numbers as Entities. International Conference on Information and Knowledge Management, Proceedings. 2022. p. 4535–4539.

Published In

International Conference on Information and Knowledge Management, Proceedings

DOI

Publication Date

October 17, 2022

Start / End Page

4535 / 4539