Research map / Label-efficient and robust learning / Recognition and long-tail
A Radical-Aware Attention-Based Model for Chinese Text Classification
Hanqing Tao, Shiwei Tong, Hongke Zhao, Tong Xu, Binbin Jin, Qi Liu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 radical features with attention and serialized BLSTMs (RAFG vs C-BLSTMs) (ablation) | RAFG · dataset#1 | F1: 0.8128 -> 0.8181 | Improved |
| add radical features with attention and serialized BLSTMs (RAFG vs C-BLSTMs) (ablation) | RAFG · dataset#2 | F1: 0.7956 -> 0.7999 | Improved |
| add radicals via four parallel BLSTMs without attention (ablation) | C-BLSTMs · dataset#1 | F1: 0.8128 -> 0.8098 | Hurt |
| add radicals via four parallel LSTMs without attention (ablation) | C-LSTMs · dataset#1 | F1: 0.8112 -> 0.8072 | Hurt |
| add attention and serialized structure over four BLSTMs (ablation) | Four BLSTMs · dataset#1 | F1: 0.8098 -> 0.8181 | Improved |
| use bidirectional LSTM instead of unidirectional (ablation) | C-LSTMs · dataset#1 | F1: 0.8112 -> 0.8128 | Improved |
| add characters to words (C-LSTMs vs word LSTM) (ablation) | LSTM (EW) · dataset#1 | F1: 0.8029 -> 0.8112 | Improved |
| use words instead of characters as LSTM input (ablation) | LSTM · dataset#1 | F1: 0.7077 -> 0.8029 | Improved |
| use word-level radicals instead of character-level radicals (ablation) | SVM + BOW · dataset#1 | F1: 0.4697 -> 0.6834 | Improved |
| RAFG vs C-BLSTMs across the 32 classes | RAFG · dataset#1 per-class | F1 | Mixed |
Closest papers by meaning
- Ideography Leads Us to the Field of Cognition: A Radical-Guided Associative Model for Chinese Text Classification (AAAI 2021 · similarity 0.87)
- MSA2: Multi-task Framework with Structure-aware and Style-adaptive Character Representation for Open-set Chinese Text Recognition (ICCV 2025 · similarity 0.84)
- An Exemplar-based Framework for Chinese Text Recognition (AAAI 2025 · similarity 0.82)
- Towards Unified Multi-granularity Text Detection with Interactive Attention (ICML 2024 · similarity 0.81)
- Language Model Meets Prototypes: Towards Interpretable Text Classification Models through Prototypical Networks (AAAI 2025 · similarity 0.81)
- McHirc: A Multimodal Benchmark for Chinese Idiom Reading Comprehension (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.
