An Empirical Study on Finding Spans
Gu, Weiwei, Zheng, Boyuan, Chen, Yunmo, Chen, Tongfei, Van Durme, Benjamin
–arXiv.org Artificial Intelligence
We present an empirical study on methods for span finding, the selection of consecutive tokens in text for some downstream tasks. We focus on approaches that can be employed in training end-to-end information extraction systems, and find there is no definitive solution without considering task properties, and provide our observations to help with future design choices: 1) a tagging approach often yields higher precision while span enumeration and boundary prediction provide higher recall; 2) span type information can benefit a boundary prediction approach; 3) additional contextualization does not help span finding in most cases.
arXiv.org Artificial Intelligence
Oct-13-2022
- Country:
- Oceania > Australia
- North America > United States
- Minnesota > Hennepin County
- Minneapolis (0.14)
- Massachusetts > Suffolk County
- Boston (0.04)
- Louisiana > Orleans Parish
- New Orleans (0.04)
- Georgia > Fulton County
- Atlanta (0.04)
- California > San Diego County
- San Diego (0.04)
- Minnesota > Hennepin County
- Europe
- France (0.04)
- Italy > Tuscany
- Florence (0.04)
- Denmark > Capital Region
- Copenhagen (0.04)
- Asia
- North Korea (0.05)
- China > Hong Kong (0.04)
- Genre:
- Research Report (0.50)
- Industry:
- Government (0.47)
- Technology: