Research map / Label-efficient and robust learning / Recognition and long-tail
Generalized test utilities for long-tail performance in extreme multi-label 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 metric-specific block-coordinate inference instead of top-k probability ranking | LightXML marginal estimates · XMLC benchmarks (Eurlex, AmazonCat, Wiki10, WikiLSHTC, Amazon-3M) | matched macro-precision, macro-recall, macro-F1, or coverage | Helped |
| Optimize a macro-measure rather than instance-wise precision@k | LightXML marginal estimates · XMLC benchmarks | instance-wise precision@k | Hurt |
| Interpolate instance-wise precision@k with macro-F1 | LightXML marginal estimates · XMLC benchmarks | instance-vs-macro performance trade-off | Mixed |
| Use propensity-scored, power-law, or log-weighted top-k inference instead of top-k probability ranking | LightXML marginal estimates · XMLC benchmarks | corresponding weighted metrics | Helped |
| Use weighted inference rather than metric-specific macro inference | LightXML marginal estimates · XMLC benchmarks | macro measures | Hurt |
| Restrict evaluation to the 1k most popular labels instead of the full label set | PLT on AmazonCat-13k · AmazonCat-13k | macro measures and coverage: decrease between 60% and 90% | Hurt |
| Restrict evaluation to the 1k most popular labels instead of the full label set | PLT on AmazonCat-13k · AmazonCat-13k | propensity-scored precision: decreases by just 1%-20% | Hurt |
Closest papers by meaning
- Harnessing Hierarchical Label Distribution Variations in Test Agnostic Long-tail Recognition (ICML 2024 · similarity 0.87)
- Consistent algorithms for multi-label classification with macro-at-$k$ metrics (ICLR 2024 · similarity 0.87)
- ELMO : Efficiency via Low-precision and Peak Memory Optimization in Large Output Spaces (ICML 2025 · similarity 0.86)
- Learning Label Shift Correction for Test-Agnostic Long-Tailed Recognition (ICML 2024 · similarity 0.86)
- Keep It on a Leash: Controllable Pseudo-label Generation Towards Realistic Long-Tailed Semi-Supervised Learning (NeurIPS 2025 · similarity 0.85)
- Long-Tailed Partial Label Learning by Head Classifier and Tail Classifier Cooperation (AAAI 2024 · similarity 0.85)
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.