Skip to main content

Learning from Multiple Noisy Augmented Data Sets for Better Cross-Lingual Spoken Language Understanding

Publication ,  Conference
Guo, Y; Shou, L; Pei, J; Gong, M; Xu, M; Wu, Z; Jiang, D
Published in: EMNLP 2021 - 2021 Conference on Empirical Methods in Natural Language Processing, Proceedings
January 1, 2021

Lack of training data presents a grand challenge to scaling out spoken language understanding (SLU) to low-resource languages. Although various data augmentation approaches have been proposed to synthesize training data in low-resource target languages, the augmented data sets are often noisy, and thus impede the performance of SLU models. In this paper we focus on mitigating noise in augmented data. We develop a denoising training approach. Multiple models are trained with data produced by various augmented methods. Those models provide supervision signals to each other. The experimental results show that our method outperforms the existing state of the art by 3.05 and 4.24 percentage points on two benchmark datasets, respectively. The code will be made open sourced on github.

Duke Scholars

Published In

EMNLP 2021 - 2021 Conference on Empirical Methods in Natural Language Processing, Proceedings

Publication Date

January 1, 2021

Start / End Page

3226 / 3237
 

Citation

APA
Chicago
ICMJE
MLA
NLM
Guo, Y., Shou, L., Pei, J., Gong, M., Xu, M., Wu, Z., & Jiang, D. (2021). Learning from Multiple Noisy Augmented Data Sets for Better Cross-Lingual Spoken Language Understanding. In EMNLP 2021 - 2021 Conference on Empirical Methods in Natural Language Processing, Proceedings (pp. 3226–3237).
Guo, Y., L. Shou, J. Pei, M. Gong, M. Xu, Z. Wu, and D. Jiang. “Learning from Multiple Noisy Augmented Data Sets for Better Cross-Lingual Spoken Language Understanding.” In EMNLP 2021 - 2021 Conference on Empirical Methods in Natural Language Processing, Proceedings, 3226–37, 2021.
Guo Y, Shou L, Pei J, Gong M, Xu M, Wu Z, et al. Learning from Multiple Noisy Augmented Data Sets for Better Cross-Lingual Spoken Language Understanding. In: EMNLP 2021 - 2021 Conference on Empirical Methods in Natural Language Processing, Proceedings. 2021. p. 3226–37.
Guo, Y., et al. “Learning from Multiple Noisy Augmented Data Sets for Better Cross-Lingual Spoken Language Understanding.” EMNLP 2021 - 2021 Conference on Empirical Methods in Natural Language Processing, Proceedings, 2021, pp. 3226–37.
Guo Y, Shou L, Pei J, Gong M, Xu M, Wu Z, Jiang D. Learning from Multiple Noisy Augmented Data Sets for Better Cross-Lingual Spoken Language Understanding. EMNLP 2021 - 2021 Conference on Empirical Methods in Natural Language Processing, Proceedings. 2021. p. 3226–3237.

Published In

EMNLP 2021 - 2021 Conference on Empirical Methods in Natural Language Processing, Proceedings

Publication Date

January 1, 2021

Start / End Page

3226 / 3237