Not logged in.

Contribution Details

Type Conference or Workshop Paper
Scope Discipline-based scholarship
Published in Proceedings Yes
Title When does aggregating multiple skills with multi-task learning work? A case study in financial NLP
Organization Unit
  • Markus Leippold
  • Jingwei Ni
  • Zhijing Jin
  • Qian Wang
  • Mrinmaya Sachan
Presentation Type paper
Item Subtype Original Work
Refereed Yes
Status Published in final form
  • English
Page Range 7465 - 7488
Event Title 61st Annual Meeting of the Association for Computational Linguistics (ACL’23)
Event Type conference
Event Location Toronto, Canada
Event Start Date July 9 - 2023
Event End Date July 14 - 2023
Series Name Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics
Number 1
Place of Publication Toronto, Canada
Publisher Association for Computational Linguistics
Abstract Text Multi-task learning (MTL) aims at achieving a better model by leveraging data and knowledge from multiple tasks. However, MTL does not always work – sometimes negative transfer occurs between tasks, especially when aggregating loosely related skills, leaving it an open question when MTL works. Previous studies show that MTL performance can be improved by algorithmic tricks. However, what tasks and skills should be included is less well explored. In this work, we conduct a case study in Financial NLP where multiple datasets exist for skills relevant to the domain, such as numeric reasoning and sentiment analysis. Due to the task difficulty and data scarcity in the Financial NLP domain, we explore when aggregating such diverse skills from multiple datasets with MTL can work. Our findings suggest that the key to MTL success lies in skill diversity, relatedness between tasks, and choice of aggregation size and shared capacity. Specifically, MTL works well when tasks are diverse but related, and when the size of the task aggregation and the shared capacity of the model are balanced to avoid overwhelming certain tasks.
Free access at Official URL
Official URL
Other Identification Number merlin-id:23709
PDF File Download from ZORA
Export BibTeX