Local Additivity Based Data Augmentation for Semi-supervised NER

Chen, Jiaao; Wang, Zhenghui; Tian, Ran; Yang, Zichao; Yang, Diyi

doi:10.18653/v1/2020.emnlp-main.95

Cited by 31 publications

(28 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They test its performance on text classification datasets. Chen et al [79] introduce Mixup into NER, proposing both Intra-LADA and InterLADA.…”

Section: Mixupmentioning

confidence: 99%

Data Augmentation Approaches in Natural Language Processing: A Survey

Li,

Hou,

Che

2021

Preprint

View full text Add to dashboard Cite

As an effective strategy, data augmentation (DA) alleviates data scarcity scenarios where deep learning techniques may fail. It is widely applied in computer vision then introduced to natural language processing and achieves improvements in many tasks. One of the main focuses of the DA methods is to improve the diversity of training data, thereby helping the model to better generalize to unseen testing data. In this survey, we frame DA methods into three categories based on the diversity of augmented data, including paraphrasing, noising, and sampling. Our paper sets out to analyze DA methods in detail according to the above categories. Further, we also introduce their applications in NLP tasks as well as the challenges.

show abstract

“…They test its performance on text classification datasets. Chen et al [79] introduce Mixup into NER, proposing both Intra-LADA and InterLADA.…”

Section: Mixupmentioning

confidence: 99%

Data Augmentation Approaches in Natural Language Processing: A Survey

Li,

Hou,

Che

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Soft data augmentation. In addition to explicit generation of concrete examples, soft augmentation, which directly represents generated examples in a continuous vector space, has been proposed: Gao et al (2019) propose to perform soft word substitution for machine translation; recent work has adapted the mix-up method (Zhang et al, 2018), which augments the original dataset by linearly interpolating the vector representations of text and labels, to text classification (Guo et al, 2019;Sun et al, 2020), named entity recognition (Chen et al, 2020) and compositional generalization (Guo et al, 2020).…”

Section: Related Workmentioning

confidence: 99%

Substructure Substitution: Structured Data Augmentation for NLP

Shi

Livescu

Gimpel

2021

Preprint

View full text Add to dashboard Cite

We study a family of data augmentation methods, substructure substitution (SUB 2 ), for natural language processing (NLP) tasks. SUB 2 generates new examples by substituting substructures (e.g., subtrees or subsequences) with ones with the same label, which can be applied to many structured NLP tasks such as part-of-speech tagging and parsing. For more general tasks (e.g., text classification) which do not have explicitly annotated substructures, we present variations of SUB 2 based on constituency parse trees, introducing structureaware data augmentation methods to general NLP tasks. For most cases, training with the augmented dataset by SUB 2 achieves better performance than training with the original training set. Further experiments show that SUB 2 has more consistent performance than other investigated augmentation methods, across different tasks and sizes of the seed dataset.

show abstract

“…To alleviate the data-sparsity issue, various advanced techniques have emerged, such as transfer learning (Pan and Yang, 2009), semi-supervised learning (Mishra and Diesner, 2016;He and Sun, 2017;Wang et al, 2020b;Bhattacharjee et al, 2020;Chen et al, 2020b), domain adaptation (Li et al, , 2019b, and data augmentation (Dai and Adel, 2020;Chen et al, 2020a;Ding et al, 2020). Considering the multilingual setting, data augmentation may be infeasible and could bring in external knowledge errors.…”

Section: Multilingual Sequence Labelingmentioning

confidence: 99%

MetaTS: Meta Teacher-Student Network for Multilingual Sequence Labeling with Minimal Supervision

Li¹,

Zhang²,

Cao³

et al. 2021

Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing

View full text Add to dashboard Cite

Sequence labeling aims to predict a finegrained sequence of labels for the text. However, such formulation hinders the effectiveness of supervised methods due to the lack of token-level annotated data. This is exacerbated when we meet a diverse range of languages. In this work, we explore multilingual sequence labeling with minimal supervision using a single unified model for multiple languages. Specifically, we propose a Meta Teacher-Student (MetaTS) Network, a novel meta learning method to alleviate data scarcity by leveraging large multilingual unlabeled data. Prior teacher-student frameworks of self-training rely on rigid teaching strategies, which may hardly produce high-quality pseudo-labels for consecutive and interdependent tokens. On the contrary, MetaTS allows the teacher to dynamically adapt its pseudoannotation strategies by the student's feedback on the generated pseudo-labeled data of each language and thus mitigate error propagation from noisy pseudo-labels. Extensive experiments on both public and real-world multilingual sequence labeling datasets empirically demonstrate the effectiveness of MetaTS 1 .

show abstract

Local Additivity Based Data Augmentation for Semi-supervised NER

Cited by 31 publications

References 28 publications

Data Augmentation Approaches in Natural Language Processing: A Survey

Data Augmentation Approaches in Natural Language Processing: A Survey

Substructure Substitution: Structured Data Augmentation for NLP

MetaTS: Meta Teacher-Student Network for Multilingual Sequence Labeling with Minimal Supervision

Contact Info

Product

Resources

About