Research map / Label-efficient and robust learning / Recognition and long-tail
Discover Multiple Novel Labels in Multi-Instance Multi-Label Learning
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 DMNL instead of MIML-NC | MIML · MSRCv2 (novel labels 1-4) | FINL: .19±.02 -> .22±.02 | Helped |
| use DMNL instead of MIML-NC | MIML · Letter Carroll (1-4) | FINL: .24±.02 -> .31±.04 | Helped |
| use DMNL instead of MIML-NC | MIML · MNIST (1-2) | FINL: .15±.05 -> .19±.04 | Helped |
| use DMNL instead of MIML-NC | MIML · Four datasets | runtime (s): 206 -> 17 (MSRCv2) | Helped |
| set number of novel labels k equal to ground truth (k=2) on toy data (ablation) | DMNL · toy data | detection performance | Helped |
| set k larger than the truth (ablation) | DMNL · toy data | detection performance | Hurt |
| set k=1 with two true novel labels (ablation) | DMNL · toy data | detection performance | Hurt |
| use DMNL instead of MIML-NC for instance detection of novel labels | MIML · real datasets | AUC | No clear effect |
| use DMNL instead of MIML-NC for known label annotation | MIML · real datasets | accuracy | No clear effect |
| use manifold structure regularizer instead of cluster structure (ablation) | DMNL · real datasets | performance | No clear effect |
Closest papers by meaning
- Scalable Optimization of Multivariate Performance Measures in Multi-instance Multi-label Learning (AAAI 2017 · similarity 0.90)
- T-MDML: Triplet-based Multiple Distance Metric Learning for Multi-Instance Multi-Label Classification with Label Correlation (AAAI 2025 · similarity 0.89)
- Multi-Instance Multi-Label Classification from Crowdsourced Labels (AAAI 2025 · similarity 0.88)
- Fast Multi-Instance Partial-Label Learning (AAAI 2025 · similarity 0.87)
- Multi-Instance Partial-Label Learning with Margin Adjustment (NeurIPS 2024 · similarity 0.86)
- Multi-View Multi-Label Classification via View-Label Matching Selection (AAAI 2025 · 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.