Abstract
Tokenization has been shown to impact the quality of downstream tasks, such as Neural Machine Translation (NMT), which is susceptible to out-of-vocabulary words and low frequency training data. Current state-of-the-art algorithms have been helpful in addressing the issues of out-of-vocabulary words, bigger vocabulary sizes and token frequency by implementing subword segmentation. We argue, however, that there is still room for improvement, in particular regarding low-frequency tokens in the training data. In this paper, we present “High Frequency Tokenizer”, or HFT, a new language-independent subword segmentation algorithm that addresses this issue. We also propose a new metric to measure the frequency coverage of a tokenizer’s vocabulary, based on a frequency rank weighted average of the frequency values of its items. We experiment with a diverse set of language corpora, vocabulary sizes, and writing systems and report improvements on both frequency statistics and on the average length of the output. We also observe a positive impact on downstream NMT.- Anthology ID:
- 2022.loresmt-1.8
- Volume:
- Proceedings of the Fifth Workshop on Technologies for Machine Translation of Low-Resource Languages (LoResMT 2022)
- Month:
- October
- Year:
- 2022
- Address:
- Gyeongju, Republic of Korea
- Venue:
- LoResMT
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 56–63
- Language:
- URL:
- https://aclanthology.org/2022.loresmt-1.8
- DOI:
- Cite (ACL):
- Edoardo Signoroni and Pavel Rychlý. 2022. HFT: High Frequency Tokens for Low-Resource NMT. In Proceedings of the Fifth Workshop on Technologies for Machine Translation of Low-Resource Languages (LoResMT 2022), pages 56–63, Gyeongju, Republic of Korea. Association for Computational Linguistics.
- Cite (Informal):
- HFT: High Frequency Tokens for Low-Resource NMT (Signoroni & Rychlý, LoResMT 2022)
- PDF:
- https://preview.aclanthology.org/ingestion-script-update/2022.loresmt-1.8.pdf
- Code
- edoardosignoroni/hftoks-eval