Abstract
The NLPCC 2021 Few-shot Learning for Chinese Language Understanding Evaluation (FewCLUE) shared task seeks for the best solution to few-shot learning tasks with pre-trained language models. This paper presents Tencent Cloud Xiaowei’s approach to this challenge, which won the 2st place in the contest. We propose a Multiple-Pattern-Exploiting Training method (MPET) for the challenge. Different from the original PET, MPET constructs multiple patterns to enhance the model’s generalization capability. We take the MPET as an auxiliary task, and jointly optimize classification and MPET. Empirical results show that our MPET is effective to few-shot learning tasks.
J. Zeng and Y. Jiang—contributed equally to this work.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
- 2.
- 3.
PET computes class probabilities using the logits that correspond to the labels for a specific task. In contrast, inspired by ADAPET [11], we computes the probability of each token in the vocabulary tokens in this paper.
- 4.
- 5.
We replace the token “#idom” in content with “[MASK]”, and make it as a regular MLM objective.
References
Bowman, S.R., Angeli, G., Potts, C., Manning, C.D.: A large annotated corpus for learning natural language inference. In: Màrquez, L., Callison-Burch, C., Su, J., Pighin, D., Marton, Y. (eds.) Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, EMNLP 2015, 17–21 September 2015, Lisbon, Portugal, pp. 632–642. The Association for Computational Linguistics (2015). https://doi.org/10.18653/v1/d15-1075
Devlin, J., Chang, M.W., Lee, K., Toutanova, K.: BERT: pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805 (2018)
Gururangan, S., et al.: Don’t stop pretraining: adapt language models to domains and tasks. arXiv preprint arXiv:2004.10964 (2020)
Hu, H., Richardson, K., Xu, L., Li, L., Kübler, S., Moss, L.S.: OCNLI: original Chinese natural language inference. In: Cohn, T., He, Y., Liu, Y. (eds.) Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: Findings, EMNLP 2020, Online Event, 16–20 November 2020. Findings of ACL, vol. EMNLP 2020, pp. 3512–3526. Association for Computational Linguistics (2020). https://doi.org/10.18653/v1/2020.findings-emnlp.314
Lai, G., Xie, Q., Liu, H., Yang, Y., Hovy, E.H.: RACE: large-scale reading comprehension dataset from examinations. In: Palmer, M., Hwa, R., Riedel, S. (eds.) Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, EMNLP 2017, 9–11 September 2017, Copenhagen, Denmark, pp. 785–794. Association for Computational Linguistics (2017). https://doi.org/10.18653/v1/d17-1082
Lan, Z., Chen, M., Goodman, S., Gimpel, K., Sharma, P., Soricut, R.: ALBERT: a lite BERT for self-supervised learning of language representations. arXiv preprint arXiv:1909.11942 (2019)
Lewis, M., et al.: BART: denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehension. arXiv preprint arXiv:1910.13461 (2019)
Liu, Y., et al.: RoBERTa: A robustly optimized BERT pretraining approach. arXiv preprint arXiv:1907.11692 (2019)
Rajpurkar, P., Zhang, J., Lopyrev, K., Liang, P.: SQuAD: 100, 000+ Questions for machine comprehension of text. In: Su, J., Carreras, X., Duh, K. (eds.) Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing, EMNLP 2016, 1–4 November 2016, Austin, Texas, USA, pp. 2383–2392. The Association for Computational Linguistics (2016). https://doi.org/10.18653/v1/d16-1264
Schick, T., Schütze, H.: Exploiting cloze-questions for few-shot text classification and natural language inference. In: Merlo, P., Tiedemann, J., Tsarfaty, R. (eds.) Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume, EACL 2021, Online, 19–23 April 2021, pp. 255–269. Association for Computational Linguistics (2021). https://www.aclweb.org/anthology/2021.eacl-main.20/
Tam, D., Menon, R.R., Bansal, M., Srivastava, S., Raffel, C.: Improving and simplifying pattern exploiting training. CoRR abs/2103.11955 (2021). https://arxiv.org/abs/2103.11955
Wang, A., Singh, A., Michael, J., Hill, F., Levy, O., Bowman, S.R.: GLUE: a multi-task benchmark and analysis platform for natural language understanding. In: 7th International Conference on Learning Representations, ICLR 2019, New Orleans, 6–9 May 2019, LA, USA. OpenReview.net (2019). https://openreview.net/forum?id=rJ4km2R5t7
Whang, T., Lee, D., Lee, C., Yang, K., Oh, D., Lim, H.: Domain adaptive training BERT for response selection. arXiv preprint arXiv:1908.04812 (2019)
Xu, L., et al.: CLUE: a Chinese language understanding evaluation benchmark. In: Scott, D., Bel, N., Zong, C. (eds.) Proceedings of the 28th International Conference on Computational Linguistics, COLING 2020, 8–13 December 2020, Barcelona, Spain (Online), pp. 4762–4772. International Committee on Computational Linguistics (2020). https://doi.org/10.18653/v1/2020.coling-main.419
Xu, L., et al.: FewCLUE: a Chinese few-shot learning evaluation benchmark. CoRR abs/2107.07498 (2021). arXiv: 2107.07498
Yang, Z., Dai, Z., Yang, Y., Carbonell, J., Salakhutdinov, R.R., Le, Q.V.: XLNet: generalized autoregressive pretraining for language understanding. In: Advances in Neural Information Processing Systems, vol. 32 (2019)
Zheng, C., Huang, M., Sun, A.: ChiD: a large-scale Chinese idiom dataset for cloze test. In: Korhonen, A., Traum, D.R., Màrquez, L. (eds.) Proceedings of the 57th Conference of the Association for Computational Linguistics, ACL 2019, 28 July–2 August 2019, Florence, Italy, vol. 1, Long Papers, pp. 778–787. Association for Computational Linguistics (2019). https://doi.org/10.18653/v1/p19-1075
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Zeng, J., Jiang, Y., Wu, S., Li, M. (2021). Enhanced Few-Shot Learning with Multiple-Pattern-Exploiting Training. In: Wang, L., Feng, Y., Hong, Y., He, R. (eds) Natural Language Processing and Chinese Computing. NLPCC 2021. Lecture Notes in Computer Science(), vol 13029. Springer, Cham. https://doi.org/10.1007/978-3-030-88483-3_31
Download citation
DOI: https://doi.org/10.1007/978-3-030-88483-3_31
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-88482-6
Online ISBN: 978-3-030-88483-3
eBook Packages: Computer ScienceComputer Science (R0)