Research map / Label-efficient and robust learning / Domain adaptation and generalization
Neural Networks Incorporating Dictionaries for Chinese Word Segmentation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add dictionary features via HyperLSTM (Model-II) | Bi-LSTM-CRF · PKU | F1: 95.1 -> 96.5 | Helped |
| add dictionary features via HyperLSTM (Model-II) | Bi-LSTM-CRF · MSR | F1: 97.0 -> 97.8 | Helped |
| add dictionary features via HyperLSTM (Model-II) | Bi-LSTM-CRF · CITYU | F1: 95.3 -> 96.3 | Helped |
| replace parallel Bi-LSTM integration (Model-I) with HyperLSTM weight generation (Model-II) (ablation) | dictionary-augmented Bi-LSTM-CRF · PKU | F1: 96.2 -> 96.5 | Helped |
| stack a second Bi-LSTM layer (ablation) | Bi-LSTM-CRF · PKU/MSR/AS/CITYU/CTB6 | F1: 95.1 -> 95.3 (PKU) | No clear effect |
| add domain-specific dictionary at test time (ablation) | Model-II · SIGHAN2010 Literature | F1: 92.87 -> 94.76 | Helped |
| add domain-specific dictionary at test time (ablation) | Model-II · SIGHAN2010 Medicine | F1: 91.27 -> 94.18 | Helped |
| add general dictionary features (no domain dict) in cross-domain setting | Bi-LSTM-CRF · SIGHAN2010 Literature | F1: 93.05 -> 92.87 (Model-II) | Hurt |
| increase dictionary size from 80% to 100% of words (ablation) | Model-II · PKU, MSR, CTB6 | F1 | Helped |
| vary HyperLSTM/feature LSTM hidden units d_ht (32 to 160) (ablation) | Model-I and Model-II · PKU, MSR, CTB6 | F1 | No clear effect |
Closest papers by meaning
- Switch-LSTMs for Multi-Criteria Chinese Word Segmentation (AAAI 2019 · similarity 0.87)
- Unsupervised Learning Helps Supervised Neural Word Segmentation (AAAI 2019 · similarity 0.86)
- BTS: A Bi-Lingual Benchmark for Text Segmentation in the Wild (CVPR 2022 · similarity 0.83)
- Morphological Segmentation with Window LSTM Neural Networks (AAAI 2016 · similarity 0.83)
- Joint Word Segmentation, POS-Tagging and Syntactic Chunking (AAAI 2016 · similarity 0.82)
- Vietnamese Words Are Not Constructed from Syllables: Rethinking the Role of Word Segmentation in Natural Language Processing for Vietnamese Texts (AAAI 2025 · similarity 0.81)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.