Research map / Label-efficient and robust learning / Recognition and long-tail
Label-Specific Feature Augmentation for Long-Tailed Multi-Label Text Classification
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 |
|---|---|---|---|
| Use LSFA instead of LDGN | MLTC · AAPD | P@5: 42.29 -> 43.43 | Helped |
| Use LSFA instead of LDGN | MLTC · RCV1 | P@5: 57.29 -> 57.52 | Helped |
| Use LSFA instead of LDGN | MLTC · EUR-Lex | P@5: 56.36 -> 58.95 | Helped |
| Use LSFA instead of HTTN | MLTC tail-labels · EUR-Lex | PSP@5: 45.74 -> 50.69 | Helped |
| Add prototypical supervised contrastive (PSC) loss (ablation) | LSFA · AAPD and EUR-Lex | P@5, N@5, PSP@5 | Helped |
| Add tail-label feature augmentation (ablation) | LSFA without PSC loss · AAPD and EUR-Lex | P@5, N@5, PSP@5 | Helped |
| Add the label-specific encoder (ablation) | LSFA without PSC loss or feature augmentation · AAPD and EUR-Lex | P@5, N@5, PSP@5 | Helped |
| Increase the head-to-tail threshold Nt (ablation) | LSFA · EUR-Lex | P@5 | Mixed |
| Increase augmentation count Na from 1 to 10 (ablation) | LSFA · EUR-Lex | P@5 | Helped |
Closest papers by meaning
- Does Head Label Help for Long-Tailed Multi-Label Text Classification (AAAI 2021 · similarity 0.89)
- Compositional Generalization for Multi-Label Text Classification: A Data-Augmentation Approach (AAAI 2024 · similarity 0.87)
- CascadeXML: Rethinking Transformers for End-to-end Multi-resolution Training in Extreme Multi-label Classification (NeurIPS 2022 · similarity 0.86)
- MLC-NC: Long-Tailed Multi-Label Image Classification Through the Lens of Neural Collapse (AAAI 2025 · similarity 0.84)
- Double-Layer Hybrid-Label Identification Feature Selection for Multi-View Multi-Label Learning (AAAI 2024 · similarity 0.84)
- Reconsidering Feature Structure Information and Latent Space Alignment in Partial Multi-label Feature Selection (AAAI 2025 · similarity 0.84)
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.