Hierarchical Latent Alignment for Non-Autoregressive Generation under High Compression Ratio-Reference-Cited by-同舟云学术

Hierarchical Latent Alignment for Non-Autoregressive Generation under High Compression Ratio

Published:2024-03-01 Issue:3 Volume:E107.D Page:411-419
ISSN:0916-8532
Container-title:IEICE Transactions on Information and Systems
language:en
Short-container-title:IEICE Trans. Inf. & Syst.

Author:

XU Wang¹,MA Yongliang²,CHEN Kehai¹,ZHOU Ming²,YANG Muyun¹,ZHAO Tiejun¹

Affiliation:

1. School of Computer Science and Technology, Harbin Institute of Technology

2. Beijing Langboat Technology Co., Ltd.

Publisher

Institute of Electronics, Information and Communications Engineers (IEICE)

Link

https://www.jstage.jst.go.jp/article/transinf/E107.D/3/E107.D_2023EDP7111/_pdf

Reference41 articles.

1. [1] J. Gu, J. Bradbury, C. Xiong, V.O. Li, and R. Socher, “Non-autoregressive neural machine translation,” International Conference on Learning Representations, 2018.

2. [2] M. Ghazvininejad, O. Levy, Y. Liu, and L. Zettlemoyer, “Mask-predict: Parallel decoding of conditional masked language models,” Proc. Conference on Empirical Methods in Natural Language Processing and the International Joint Conference on Natural Language Processing, pp.6112-6121, 2019. 10.18653/v1/d19-1633

3. [3] J. Guo, X. Tan, L. Xu, T. Qin, E. Chen, and T.-Y. Liu, “Fine-tuning by curriculum learning for non-autoregressive neural machine translation,” Proc. AAAI Conference on Artificial Intelligence, pp.7839-7846, 2020. 10.1609/aaai.v34i05.6289

4. [4] W. Qi, Y. Gong, J. Jiao, Y. Yan, W. Chen, D. Liu, K. Tang, H. Li, J. Chen, R. Zhang, M. Zhou, and N. Duan, “Bang: Bridging autoregressive and non-autoregressive generation with large scale pretraining,” Proc. International Conference on Machine Learning, pp.8630-8639, 2021.

5. [5] P. Liu, C. Huang, and L. Mou, “Learning non-autoregressive models from search for unsupervised sentence summarization,” Proc. Annual Meeting of the Association for Computational Linguistics, pp.7916-7929, 2022. 10.18653/v1/2022.acl-long.545