Global Pointer: Novel Efficient Span-based Approach for Named Entity Recognition

@article{Su2022GlobalPN,
  title={Global Pointer: Novel Efficient Span-based Approach for Named Entity Recognition},
  author={Jianlin Su and Ahmed Murtadha and Shengfeng Pan and Jing Hou and Jun Sun and Wanwei Huang and Bo Wen and Yunfeng Liu},
  journal={ArXiv},
  year={2022},
  volume={abs/2208.03054}
}
We extensively evaluate GP on various benchmark datasets. Our extensive experiments demonstrate that GP can outperform the existing solution. Moreover, the experimental results show the efficacy of the introduced loss function compared to softmax and entropy alternatives. experimental results show the efficacy of the introduced loss function compared to softmax and entropy alternatives. 

Figures and Tables from this paper

Universal Information Extraction as Unified Semantic Matching

This paper proposes to universally model various IE tasks with a USM framework, which introduces three token linking operations to model the abilities of structuring and conceptualizing, and shows strong generalization ability in zero/few-shot transfer settings.

Named Entity Recognition of Benefit Enterprise Policy Based on RoBERTa_wwm_ext-BiLSTM-CRF

The results show that the model is better than the comparison model in terms of accuracy, recall and F1, and the F1 value reaches 86.4%, and problems such as word polysemy can be well represented.

A Token-pair Framework for Information Extraction from Dialog Transcripts in SereTOD Challenge

A token-pair framework to simultaneously identify entity and value mentions and link them into corresponding triples and significantly outperforms the baseline solution and ranks first in triple f1 and third in entity f1.

A GlobalPointer based Robust Approach for Information Extraction from Dialog Transcripts

  • Yanbo J. WangSheng Chen Xiaochen Cai
  • Computer Science
    SERETOD
  • 2022
This paper proposed a multi-model fusion approach based on GlobalPointer, combined with some optimisation tricks, and finally achieved an entity F1 of 60.73, an entity-slot-value triple F 1 of 56, and an average F2 of 58.37, which got the highest score in SereTOD 2022 Workshop challenge.

References

SHOWING 1-10 OF 67 REFERENCES

Code-Switched Named Entity Recognition with Embedding Attention

We describe our work for the CALCS 2018 shared task on named entity recognition on code-switched data. Our system ranked first place for MS Arabic-Egyptian named entity recognition and third place

Towards Improving Neural Named Entity Recognition with Gazetteers

It is shown that properly utilizing external gazetteers could benefit segmental neural NER models by adding a simple module on the recently proposed hybrid semi-Markov CRF architecture and observing some promising results.

A Boundary-aware Neural Model for Nested Named Entity Recognition

This work proposes a boundary-aware neural model for nested NER which leverages entity boundaries to predict entity categorical labels, which can decrease computation cost and relieve error propagation problem in layered sequence labeling model.

Robust Lexical Features for Improved Neural Network Named-Entity Recognition

This work proposes to embed words and entity types into a low-dimensional vector space the authors train from annotated data produced by distant supervision thanks to Wikipedia, and compute a feature vector representing each word that establishes a new state-of-the-art F1 score.

Neural Metric Learning for Fast End-to-End Relation Extraction

A novel neural architecture utilizing the table structure, based on repeated applications of 2D convolutions for pooling local dependency and metric-based features, that improves on the state-of-the-art without the need for global optimization is introduced.

Named entity recognition in query

Experimental results show that the proposed method based on WS-LDA can accurately perform NERQ, and outperform the baseline methods.

Joint Extraction of Entities and Relations Based on a Novel Tagging Scheme

A novel tagging scheme is proposed that can convert the joint extraction task to a tagging problem, and different end-to-end models are studied to extract entities and their relations directly, without identifying entities and relations separately.

Two Are Better than One: Joint Entity and Relation Extraction with Table-Sequence Encoders

It is argued that it can be beneficial to design two distinct encoders to capture such two different types of information in the learning process, and proposed is the novel {\em table-sequence encoder} where two different encoder -- a table encoder and a sequence encoder are designed to help each other in the representation learning process.

Boundary Enhanced Neural Span Classification for Nested Named Entity Recognition

This work proposes a boundary enhanced neural span classification model that has the ability to generate high-quality candidate spans and greatly reduces the time complexity during inference, and incorporates an additional boundary detection task to predict those words that are boundaries of entities.

Joint Extraction of Multiple Relations and Entities by Using a Hybrid Neural Network

This paper proposes a novel end-to-end neural model to jointly extract entities and relations in a sentence that uses a hybrid neural network to automatically learn sentence features and does not rely on any Natural Language Processing tools, such as dependency parser.
...