2021
DOI: 10.48550/arxiv.2109.07589
|View full text |Cite
Preprint
|
Sign up to set email alerts
|

CONTaiNER: Few-Shot Named Entity Recognition via Contrastive Learning

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1

Citation Types

0
3
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
2

Relationship

0
2

Authors

Journals

citations
Cited by 2 publications
(3 citation statements)
references
References 0 publications
0
3
0
Order By: Relevance
“…Contrastive Learning Contrastive learning is used in computer vision for image classification (Chen et al 2020;He et al 2020;Khosla et al 2020) and is now widely used in various tasks (Chuang et al 2020;Giorgi et al 2020;Hou et al 2020;Gao, Yao, and Chen 2021;Das et al 2021;Chen et al 2022). In NLP, (Giorgi et al 2020;Gao, Yao, and Chen 2021;Chen et al 2022) propose to enhance semantic representation and a pre-trained model based on contrastive learning.…”
Section: Related Workmentioning
confidence: 99%
See 1 more Smart Citation
“…Contrastive Learning Contrastive learning is used in computer vision for image classification (Chen et al 2020;He et al 2020;Khosla et al 2020) and is now widely used in various tasks (Chuang et al 2020;Giorgi et al 2020;Hou et al 2020;Gao, Yao, and Chen 2021;Das et al 2021;Chen et al 2022). In NLP, (Giorgi et al 2020;Gao, Yao, and Chen 2021;Chen et al 2022) propose to enhance semantic representation and a pre-trained model based on contrastive learning.…”
Section: Related Workmentioning
confidence: 99%
“…In NLP, (Giorgi et al 2020;Gao, Yao, and Chen 2021;Chen et al 2022) propose to enhance semantic representation and a pre-trained model based on contrastive learning. Hou et al (2020) apply contrastive learning for slot filling and Das et al (2021) propose CONTaiNER for few-shot NER combining contrastive learning with Gaussian distribution. Contrastive learning can effectively pull the distance between positive samples and push the distance between negative samples to achieve better recognition results.…”
Section: Related Workmentioning
confidence: 99%
“…Yang and Katiyar [12] demonstrated impressive results for standard few-shot learning applications with a model using nearest neighbor learning and structured inference. Das et al [13] proposed CONTaiNER, a new contrastive learning method that enhances the inter-token distribution distance for using the few-shot setting on NER. FewNER made use of a meta-learning technique, and it applied novel N-way K-shot learning approach for applying few-shot learning to NER [14].Fritzler, Logacheva, and Kretov [15] used a metric learning approach called Prototypical Networking which learnt the intermediate representations of tokens that fall under the same named entity category.…”
Section: Literature Reviewmentioning
confidence: 99%