Research map / Label-efficient and robust learning / Semi-supervised learning
Joint Word Segmentation, POS-Tagging and Syntactic Chunking
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 |
|---|---|---|---|
| perform segmentation, POS-tagging and chunking jointly instead of pipeline (ablation) | Pipeline chunker · CTB4 test | chunking F1: 69.02 -> 70.96 | Helped |
| perform segmentation, POS-tagging and chunking jointly instead of pipeline (ablation) | Pipeline chunker · CTB4 test | segmentation F1: 88.81 -> 89.85 | Helped |
| add semi-supervised chunk cluster and bigram features (Semi-ALL) (ablation) | Joint model · CTB4 test | chunking F1: 70.96 -> 72.29 | Helped |
| add semi-supervised chunk cluster and bigram features (Semi-ALL) (ablation) | Joint model · CTB4 test | segmentation F1: 89.85 -> 91.00 | Helped |
| add full chunk features without semi-supervision (ablation) | Joint model · CTB4 test | chunking F1: 70.96 -> 70.63 | Hurt |
| add full chunk features without semi-supervision (ablation) | Pipeline · CTB4 test | chunking F1: 69.02 -> 68.82 | Hurt |
| derive chunk features from semi-supervised cluster sets (Semi-C) (ablation) | Joint model with full chunk features · CTB4 test | chunking F1: 70.63 -> 72.09 | Helped |
| apply semi-supervised chunk cluster features (ablation) | Pipeline · CTB4 test | chunking F1: 69.02 -> 69.45 | Helped |
| use only word bigram semi-supervised feature (Semi-W) (ablation) | Joint model · CTB4 test | chunking F1: 70.96 -> 71.62 | Helped |
| deque size 4 instead of 1 (ablation) | Joint character-level chunker · CTB4 dev | chunking F1: 83.77 -> 84.52 | Helped |
| deque size 3 instead of 1 (ablation) | Joint character-level chunker · CTB4 dev | chunking F1: 83.77 -> 83.30 | Hurt |
Closest papers by meaning
- BTS: A Bi-Lingual Benchmark for Text Segmentation in the Wild (CVPR 2022 · similarity 0.86)
- Unsupervised Learning Helps Supervised Neural Word Segmentation (AAAI 2019 · similarity 0.85)
- Understanding Multi-Granularity for Open-Vocabulary Part Segmentation (NeurIPS 2024 · similarity 0.84)
- Exploring Simple Open-Vocabulary Semantic Segmentation (CVPR 2025 · similarity 0.83)
- CAT-Seg: Cost Aggregation for Open-Vocabulary Semantic Segmentation (CVPR 2024 · similarity 0.83)
- Scribbles for All: Benchmarking Scribble Supervised Segmentation Across Datasets (NeurIPS 2024 · similarity 0.83)
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.