An empirical study on finding spans
arXiv preprint arXiv:2210.06824, 2022•arxiv.org
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 …
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 …
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
以上显示的是最相近的搜索结果。 查看全部搜索结果