Abstract
As manually labelling data can be error-prone and labour-intensive, some recent studies automatically classify documents without any training on labelled data and directly exploit pre-trained language models (PLMs) for many downstream tasks, also known as zero-shot text classification. In the same vein, we propose a novel framework aims at improving zero-short learning and enriching domain specific information required by PLMs with transformer models. To unleash the power of PLMs pre-trained on massive cross-section corpus, the framework unifies two transformers for different purposes: 1) expanding categorical labels required by PLMs by creating coherent representative samples with GPT2, which is a language model acclaimed for generating sensical text outputs, and 2) augmenting documents with T5, which has the virtue of synthesizing high quality new samples similar to the original text. The proposed framework can be easily integrated into different general testbeds. Extensive experiments on two popular topic classification datasets have proved its effectiveness.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
\(\mathcal {L}\) may not be pre-defined in practical scenarios, while in the experiments we fix it for convenient evaluations.
- 2.
In this study, we use the publicly available BERT of uncased version https://huggingface.co/bert-base-uncased. The output of BERT’s NSP has two logits: the first is the probability of IsNext and the second is the probability of NotNext, both of which are outputs of the SoftMax function from the previous layer.
- 3.
- 4.
- 5.
- 6.
- 7.
- 8.
- 9.
References
Chang, M.W., Ratinov, L.A., Roth, D., Srikumar, V.: Importance of semantic representation: dataless classification. In: AAAI2008 Proceedings of the 23rd National Conference on Artificial Intelligence, vol. 2, pp. 830–835 (2008)
Chen, X., Xia, Y., Jin, P., Carroll, J.: Dataless text classification with descriptive IDA. In: AAAI2015 Proceedings of the Twenty-Ninth AAAI Conference on Artificial Intelligence, vol. 29, pp. 2224–2231 (2015)
Conneau, A., et al.: XNLI: Evaluating cross-lingual sentence representations. In: Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, EMNLP (2018)
Dathathri, S., et al.: Plug and play language models: a simple approach to controlled text generation. CoRR abs/1912.02164 (2019)
Devlin, J., Chang, M.W., Lee, K., Toutanova, K.: BERT: pre-training of deep bidirectional transformers for language understanding. In: Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2019, Minneapolis, MN, USA, pp. 4171–4186 (2019)
Elhoseiny, M., Saleh, B., Elgammal, A.: Write a classifier: zero-shot learning using purely textual descriptions. In: 2013 IEEE International Conference on Computer Vision, ICCV, pp. 2584–2591 (2013)
Gabrilovich, E., et al.: Computing semantic relatedness using Wikipedia-based explicit semantic analysis. In: IJCAI2007 Proceedings of the 20th International Joint Conference on Artifical Intelligence, IJCAI, vol. 7, pp. 1606–1611 (2007)
Harrando, I., Troncy, R.: Explainable zero-shot topic extraction using a common-sense knowledge graph. In: Third Biennial Conference on Language, Data and Knowledge, LDK (2021)
Holtzman, A., Buys, J., Du, L., Forbes, M., Choi, Y.: The curious case of neural text degeneration. CoRR abs/1904.09751 (2019)
Keskar, N.S., McCann, B., Varshney, L., Xiong, C., Socher, R.: CTRL - a conditional transformer language model for controllable generation. CoRR abs/1909.05858 (2019)
Kumar, V., Choudhary, A., Cho, E.: Data augmentation using pre-trained transformer models. CoRR abs/2003.02245 (2020)
Li, C., Xing, J., Sun, A., Ma, Z.: Effective document labeling with very few seed words: a topic model approach. In: CIKM 2016 Proceedings of the 25th ACM International on Conference on Information and Knowledge Management, CIKM, pp. 85–94 (2016)
Li, Y., Zheng, R., Tian, T., Hu, Z., Iyer, R., Sycara, K.: Joint embedding of hierarchical categories and entities for concept categorization and dataless classification. CoRR abs/1607.07956 (2016)
Liu, P., Yuan, W., Fu, J., Jiang, Z., Hayashi, H., Neubig, G.: Pre-train, prompt, and predict: a systematic survey of prompting methods in natural language processing. CoRR abs/2107.13586 (2021)
Liu, Y., et al.: RoBERTa: a robustly optimized BERT pretraining approach. CoRR abs/1907.11692 (2019)
Longpre, S., Wang, Y., DuBois, C.: How effective is task-agnostic data augmentation for pretrained transformers? In: the 2020 Conference on Empirical Methods in Natural Language Processing, EMNLP, pp. 4401–4411 (2020)
Nie, Y., Williams, A., Dinan, E., Bansal, M., Weston, J., Kiela, D.: Adversarial NLI: a new benchmark for natural language understanding. In: Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, ACL (2020)
Palatucci, M.M., Pomerleau, D.A., Hinton, G.E., Mitchell, T.: Zero-shot learning with semantic output codes. In: NIPS2009 Proceedings of the 22nd International Conference on Neural Information Processing Systems, NIPS, vol. 22 (2009)
Rei, M., Søgaard, A.: Zero-shot sequence labeling: transferring knowledge from sentences to tokens. In: Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers), NAACL-HLT, pp. 293–302 (2018)
Reimers, N., Gurevych, I.: Sentence-BERT: Sentence embeddings using Siamese BERT-networks. In: Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, EMNLP-IJCNLP, pp. 3982–3992 (2019)
Rios, A., Kavuluru, R.: Few-shot and zero-shot multi-label learning for structured label spaces. In: Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, EMNLP, pp. 3132–3142 (2018)
Schick, T., Schütze, H.: Exploiting cloze questions for few-shot text classification and natural language inference. Computing Research Repository arXiv:2001.07676 (2020)
Sennrich, R., Haddow, B., Birch, A.: Improving neural machine translation models with monolingual data. In: Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), ACL, pp. 86–96 (2016)
Shleifer, S.: Low resource text classification with ulmfit and backtranslation. CoRR abs/1903.09244 (2019)
Song, Y., Roth, D.: On dataless hierarchical text classification. In: AAAI2014 Proceedings of the Twenty-Eighth AAAI Conference on Artificial Intelligence, AAAI, pp. 1579–1585 (2014)
Song, Y., Upadhyay, S., Peng, H., Roth, D.: Cross-lingual dataless classification for many languages. In: Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence, IJCAI, pp. 2901–2907 (2016)
Sun, Y., Zheng, Y., Hao, C., Qiu, H.: NSP-BERT: a prompt-based zero-shot learner through an original pre-training task-next sentence prediction. arXiv:2109.03564 (2021)
Veeranna, S.P., Nam, J., Mencía, E., Furnkranz, J.: Using semantic similarity for multi-label zero-shot classification of text documents. In: European Symposium on Artificial Neural Networks, ESANN, pp. 423–428 (2016)
Wei, J., Zou, K.: Eda: Easy data augmentation techniques for boosting performance on text classification tasks. In: Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, EMNLP-IJCNLP, pp. 6383–6389 (2019)
Xie, Q., Dai, Z., Hovy, E., Luong, T., Le, Q.: Unsupervised data augmentation for consistency training. In: Advances in Neural Information Processing Systems 33 (2020)
Yang, Z., Yang, Y., Cer, D., Law, J., Darve, E.: Universal sentence representation learning with conditional masked language model. In: Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, EMNLP, pp. 6216–6228 (2021)
Yin, W., Hay, J., Roth, D.: Benchmarking zero-shot text classification: datasets, evaluation and entailment approach. In: Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, EMNLP-IJCNLP, pp. 3905–3914 (2019)
Zhang, J., Lertvittayakumjorn, P., Guo, Y.: Integrating semantic knowledge to tackle zero-shot text classification. In: Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT, pp. 1031–1040 (2019)
Zhang, X., Zhao, J., LeCun, Y.: Character-level convolutional networks for text classification. In: NIPS2015 Proceedings of the 28th International Conference on Neural Information Processing Systems, NIPS, vol. 1, pp. 649–657 (2015)
Acknowledgement
This work was supported by the NSERC Discovery Grants.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 Springer Nature Switzerland AG
About this paper
Cite this paper
Chen, Y., Liu, Y. (2022). A Novel Hybrid Framework to Enhance Zero-shot Classification. In: Sojka, P., Horák, A., Kopeček, I., Pala, K. (eds) Text, Speech, and Dialogue. TSD 2022. Lecture Notes in Computer Science(), vol 13502. Springer, Cham. https://doi.org/10.1007/978-3-031-16270-1_17
Download citation
DOI: https://doi.org/10.1007/978-3-031-16270-1_17
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-16269-5
Online ISBN: 978-3-031-16270-1
eBook Packages: Computer ScienceComputer Science (R0)