Research map / Label-efficient and robust learning / Recognition and long-tail
Extreme Multi-label Classification from Aggregated Labels
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 |
|---|---|---|---|
| Learn label embeddings and assign aggregated labels to samples, rather than using Parabel without assignment | EAGLE with Parabel · EurLex-4K, R-4 | P@1: 76.58 -> 80.87 | Helped |
| Learn label embeddings and assign aggregated labels to samples, rather than using Parabel without assignment | EAGLE with Parabel · Wiki-10K, R-4 | P@1: 73.07 -> 80.38 | Helped |
| Learn label embeddings and assign aggregated labels to samples, rather than using Parabel without assignment | EAGLE with Parabel · AmazonCat-13K, R-4 | P@1: 80.62 -> 81.34 | Helped |
| Learn label embeddings and assign aggregated labels to samples, rather than using Parabel without assignment | EAGLE with Parabel · Wiki-325K, R-4 | P@1: 34.09 -> 60.60 | Helped |
| Add iterative label learning to EAGLE-0 (ablation) | EAGLE with Parabel · EurLex-4K, R-4 | P@1: 80.16 -> 80.87 | Helped |
| Add iterative label learning to EAGLE-0 (ablation) | EAGLE with Parabel · Wiki-10K, R-4 | P@1: 78.28 -> 80.38 | Helped |
| Add iterative label learning to EAGLE-0 (ablation) | EAGLE with Parabel · AmazonCat-13K, R-4 | P@1: 78.83 -> 81.34 | Helped |
| Add iterative label learning to EAGLE-0 (ablation) | EAGLE with Parabel · Wiki-325K, R-4 | P@1: 62.45 -> 60.60 | Hurt |
| Add label embeddings and the assignment-based co-attention extension | Deep-MIML · MNIST, group size 50 | accuracy: 33.33 -> 38.46 | Helped |
| Add label embeddings and the assignment-based co-attention extension | Deep-MIML · Fashion-MNIST, group size 50 | accuracy: 19.00 -> 28.65 | Helped |
| Add label embeddings and the assignment-based co-attention extension | Deep-MIML · Yelp, group size 4 | top-1 precision: 40.69 -> 46.25 | Helped |
| Add iterative label learning to EAGLE-0 (ablation) | EAGLE MIML extension · MNIST, group size 4 | accuracy: 94.82 -> 94.82 | No clear effect |
Closest papers by meaning
- Label Disentanglement in Partition-based Extreme Multilabel Classification (NeurIPS 2021 · similarity 0.91)
- Multi-Instance Multi-Label Classification from Crowdsourced Labels (AAAI 2025 · similarity 0.88)
- Taming the Sigmoid Bottleneck: Provably Argmaxable Sparse Multi-Label Classification (AAAI 2024 · similarity 0.88)
- T-MDML: Triplet-based Multiple Distance Metric Learning for Multi-Instance Multi-Label Classification with Label Correlation (AAAI 2025 · similarity 0.88)
- Toward Robustness in Multi-Label Classification: A Data Augmentation Strategy against Imbalance and Noise (AAAI 2024 · similarity 0.87)
- ELMO : Efficiency via Low-precision and Peak Memory Optimization in Large Output Spaces (ICML 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.