Abstract
Leveled reading (LR) aims to automatically classify texts by the cognitive levels of readers, which is fundamental in providing appropriate reading materials regarding different reading capabilities. However, most state-of-the-art LR methods rely on the availability of copious annotated resources, which prevents their adaptation to low-resource languages like Chinese. In our work, to tackle LR in Chinese, we explore how different language transfer methods perform on English-Chinese LR. Specifically, we focus on adversarial training and cross-lingual pre-training method to transfer the LR knowledge learned from annotated data in the resource-rich English language to Chinese. For evaluation, we first introduce the age-based standard to align datasets with different leveling standards. Then we conduct experiments in both zero-shot and few-shot settings. Comparing these two methods, quantitative and qualitative evaluations show that the cross-lingual pre-training method effectively captures the language-invariant features between English and Chinese. We conduct analysis to propose further improvement in cross-lingual LR.- Anthology ID:
- 2021.findings-emnlp.227
- Volume:
- Findings of the Association for Computational Linguistics: EMNLP 2021
- Month:
- November
- Year:
- 2021
- Address:
- Punta Cana, Dominican Republic
- Editors:
- Marie-Francine Moens, Xuanjing Huang, Lucia Specia, Scott Wen-tau Yih
- Venue:
- Findings
- SIG:
- SIGDAT
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 2677–2682
- Language:
- URL:
- https://preview.aclanthology.org/icon-24-ingestion/2021.findings-emnlp.227/
- DOI:
- 10.18653/v1/2021.findings-emnlp.227
- Cite (ACL):
- Simin Rao, Hua Zheng, and Sujian Li. 2021. Cross-Lingual Leveled Reading Based on Language-Invariant Features. In Findings of the Association for Computational Linguistics: EMNLP 2021, pages 2677–2682, Punta Cana, Dominican Republic. Association for Computational Linguistics.
- Cite (Informal):
- Cross-Lingual Leveled Reading Based on Language-Invariant Features (Rao et al., Findings 2021)
- PDF:
- https://preview.aclanthology.org/icon-24-ingestion/2021.findings-emnlp.227.pdf