Research map / Label-efficient and robust learning / Noisy labels
Label Confusion Learning to Enhance Text Classification Models
Biyang Guo, Songqiao Han, Xiao Han, Hailiang Huang, Ting Lu
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 LCM to LSTM-rand | LSTM-rand · 20NG, AG's News, DBPedia, FDCNews, THUCNews | accuracy: 4.20%, 0.43%, 0.40%, 2.62%, 0.51% | Improved |
| add LCM to LSTM-pre | LSTM-pre · AG's News | accuracy: -0.03% | No clear effect |
| add LCM to CNN-pre | CNN-pre · 20NG and FDCNews | accuracy: -0.43%, -0.15% | Hurt |
| add LCM to BERT | BERT · 20NG, AG's News, DBPedia, FDCNews, THUCNews | accuracy: 0.43%, 0.18%, 0.12%, 0.56%, 0.12% | Improved |
| use LCM instead of no enhancement | LSTM · 20NG with 6%, 12%, and 30% noise | accuracy: 0.5946 -> 0.6572; 0.5747 -> 0.6385; 0.4681 -> 0.5178 | Improved |
| use LCM rather than label smoothing | LSTM · 20NG with 6%, 12%, and 30% noise | accuracy: 0.6155 -> 0.6572; 0.6010 -> 0.6385; 0.4994 -> 0.5178 | Improved |
| use LCM rather than label smoothing | BERT · 20NG with 6%, 12%, and 30% noise | accuracy: 0.8742 -> 0.8789; 0.8535 -> 0.8581; 0.7932 -> 0.7980 | Improved |
| add LCM | LSTM basic predictor · 8NG-H | accuracy: 0.7751 -> 0.7973 | Improved |
| add LCM | LSTM basic predictor · 8NG-E | accuracy: 0.8291 -> 0.8362 | Improved |
| add LCM | LSTM basic predictor · 4NG-H | accuracy: 0.9048 -> 0.9147 | Improved |
| add LCM | LSTM basic predictor · 4NG-E | accuracy: 0.9277 -> 0.9254 | Hurt |
| add LCM to CNN basic predictor | CNN · MNIST and Fashion MNIST | accuracy: 0.9822 -> 0.9841; 0.8929 -> 0.9028 | Improved |
Closest papers by meaning
- Knowledge Amalgamation for Multi-Label Classification via Label Dependency Transfer (AAAI 2023 · similarity 0.85)
- Collaborative Learning of Label Semantics and Deep Label-Specific Features for Multi-Label Classification (TPAMI 2021 · similarity 0.84)
- Compositional Generalization for Multi-Label Text Classification: A Data-Augmentation Approach (AAAI 2024 · similarity 0.84)
- Distilling Reliable Knowledge for Instance-Dependent Partial Label Learning (AAAI 2024 · similarity 0.84)
- Adaptive Integration of Partial Label Learning and Negative Learning for Enhanced Noisy Label Learning (AAAI 2024 · similarity 0.83)
- Combating Semantic Contamination in Learning with Label Noise (AAAI 2025 · 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.
