# Chinese Lexical Analysis with Deep Bi-GRU-CRF Network

arXiv: https://arxiv.org/abs/1807.01882 · HTML: https://mlsanity.com/paper/1807.01882/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 5 Jul 2018.

## Built on by (most cited first, 9 of 9)

- [Integrating Pattern- and Fact-based Fake News Detection via Model Preference Learning](https://mlsanity.com/paper/2109.11333/) (2021)
- [SeACo-Paraformer: A Non-Autoregressive ASR System with Flexible and Effective Hotword Customization Ability](https://mlsanity.com/paper/2308.03266/) (2023)
- [Less Learn Shortcut: Analyzing and Mitigating Learning of Spurious Feature-Label Correlation](https://mlsanity.com/paper/2205.12593/) (2022)
- [CSCD-NS: a Chinese Spelling Check Dataset for Native Speakers](https://mlsanity.com/paper/2211.08788/) (2022)
- [A Hong Kong Sign Language Corpus Collected from Sign-interpreted TV News](https://mlsanity.com/paper/2405.00980/) (2024)
- [FDLLM: A Dedicated Detector for Black-Box LLMs Fingerprinting](https://mlsanity.com/paper/2501.16029/) (2025)
- [Tgea: An error-annotated dataset and benchmark tasks for text generation from pretrained language models](https://mlsanity.com/paper/2503.04232/) (2025)
- [Exploiting Word Semantics to Enrich Character Representations of Chinese Pre-trained Models](https://mlsanity.com/paper/2207.05928/) (2022)
- [Eval-GCSC: A New Metric for Evaluating ChatGPT's Performance in Chinese Spelling Correction](https://mlsanity.com/paper/2311.08219/) (2023)

## Abstract (arXiv)

Lexical analysis is believed to be a crucial step towards natural language understanding and has been widely studied. Recent years, end-to-end lexical analysis models with recurrent neural networks have gained increasing attention. In this report, we introduce a deep Bi-GRU-CRF network that jointly models word segmentation, part-of-speech tagging and named entity recognition tasks. We trained the model using several massive corpus pre-tagged by our best Chinese lexical analysis tool, together with a small, yet high-quality human annotated corpus. We conducted balanced sampling between different corpora to guarantee the influence of human annotations, and fine-tune the CRF decoding layer regularly during the training progress. As evaluated by linguistic experts, the model achieved a 95.5% accuracy on the test set, roughly 13% relative error reduction over our (previously) best Chinese lexical analysis tool. The model is computationally efficient, achieving the speed of 2.3K characters per second with one thread.