# State-of-the-art Chinese Word Segmentation with Bi-LSTMs

arXiv: https://arxiv.org/abs/1808.06511 · HTML: https://mlsanity.com/paper/1808.06511/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 Aug 2018.

## Built on by (most cited first, 10 of 10)

- [Lexicon Enhanced Chinese Sequence Labeling Using BERT Adapter](https://mlsanity.com/paper/2105.07148/) (2021)
- [Context-Aware Legal Citation Recommendation using Deep Learning](https://mlsanity.com/paper/2106.10776/) (2021)
- [Breaking the Representation Bottleneck of Chinese Characters: Neural Machine Translation with Stroke Sequence Modeling](https://mlsanity.com/paper/2211.12781/) (2022)
- [DyLex: Incorporating Dynamic Lexicons into BERT for Sequence Labeling](https://mlsanity.com/paper/2109.08818/) (2021)
- [Span Labeling Approach for Vietnamese and Chinese Word Segmentation](https://mlsanity.com/paper/2110.00156/) (2021)
- [Chinese Word Segmentation with Heterogeneous Graph Neural Network](https://mlsanity.com/paper/2201.08975/) (2022)
- [Morphological Analysis of Japanese Hiragana Sentences using the BI-LSTM CRF Model](https://mlsanity.com/paper/2201.03366/) (2022)
- [Mining Word Boundaries in Speech as Naturally Annotated Word Segmentation Data](https://mlsanity.com/paper/2210.17122/) (2022)
- [Downstream Task-Oriented Neural Tokenizer Optimization with Vocabulary Restriction as Post Processing](https://mlsanity.com/paper/2304.10808/) (2023)
- [Tangut Word Segmentation under Extreme Resource Scarcity: Integrating Traditional Lexicons and Unlabeled Text](https://mlsanity.com/paper/2608.18437/) (2026)

## Abstract (arXiv)

A wide variety of neural-network architectures have been proposed for the task of Chinese word segmentation. Surprisingly, we find that a bidirectional LSTM model, when combined with standard deep learning techniques and best practices, can achieve better accuracy on many of the popular datasets as compared to models based on more complex neural-network architectures. Furthermore, our error analysis shows that out-of-vocabulary words remain challenging for neural-network models, and many of the remaining errors are unlikely to be fixed through architecture changes. Instead, more effort should be made on exploring resources for further improvement.