Research map / Label-efficient and robust learning / Recognition and long-tail
Scalable Generative Models for Multi-label Learning with Missing Labels
Vikas Jain, Nirbhay Modhe, Piyush Rai
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 |
|---|---|---|---|
| turn on label exposure model (ablation) | GenEML · Synthetic multi-label data | Precision@1: 79.2 -> 87.8 | Improved |
| turn on label exposure model (ablation) | GenEML · Synthetic multi-label data | Precision@3: 59.0 -> 65.2 | Improved |
| turn on label exposure model (ablation) | GenEML · Synthetic multi-label data | Precision@2: 68.9 -> 75.2 | Improved |
| model label exposure versus LEML without exposure | GenEML · Synthetic multi-label data | Precision@1: 78.8 -> 87.8 | Improved |
| GenEML versus best baseline | GenEML · Bibtex | Precision@1: 63.46 -> 64.09 | Improved |
| GenEML versus best baseline | GenEML · Mediamill | Precision@1: 84.22 -> 87.15 | Improved |
| GenEML versus best baseline | GenEML · Eurlex-4K | Precision@1: 76.43 -> 77.75 | Improved |
| GenEML versus LEML | GenEML · Movielens | Precision@1: 54.22 -> 55.78 | Improved |
| GenEML (online EM) versus LEML on large-scale RCV | GenEML · RCV-2K | Precision@1: 89.09 -> 87.21 | Hurt |
| GenEML (online EM) versus PfasterXML on Wiki10-31K | GenEML · Wiki10-31K | Precision@5: 59.10 -> 32.06 | Hurt |
| online EM instead of batch EM | GenEML · Bibtex and Mediamill | convergence time | Improved |
| use first-order gradient updates for W instead of CG on very large data | GenEML online EM · large-scale data | memory and speed | Improved |
Closest papers by meaning
- Deep Generative Models for Weakly-Supervised Multi-Label Classification (ECCV 2018 · similarity 0.89)
- Learning Compact Semantic Information for Incomplete Multi-View Missing Multi-Label Classification (ICML 2025 · similarity 0.87)
- Incomplete Multi-View Multi-label Learning via Disentangled Representation and Label Semantic Embedding (CVPR 2025 · similarity 0.87)
- Imprecise Label Learning: A Unified Framework for Learning with Various Imprecise Label Configurations (NeurIPS 2024 · similarity 0.87)
- Attention-Induced Embedding Imputation for Incomplete Multi-View Partial Multi-Label Classification (AAAI 2024 · similarity 0.86)
- View-Category Interactive Sharing Transformer for Incomplete Multi-View Multi-Label Learning (CVPR 2024 · similarity 0.86)
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.
