Research map / Label-efficient and robust learning / Recognition and long-tail
Rethinking Classifier Re-Training in Long-Tailed Recognition: Label Over-Smooth Can Balance
Siyu Sun, Han Lu, Jiangtong Li, Yichen Xie, Tianjiao Li, Xiaokang Yang, Liqing Zhang, Junchi Yan
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Studies classifier re-training in decoupled long-tailed recognition, isolating it from representation learning, and proposes label over-smoothing as a way to balance the classifier.
Problem
Decoupled training for long-tailed recognition improves both representation and classifier stages at once, which obscures the effect of classifier re-training.
Approach
Fix a strong representation learned with simple regularization and study classifier re-training alone; the paper proposes label over-smoothing as a mechanism that can balance the classifier. Full method details are not available from the abstract.
Experimental setup
Builds on: decoupled training
Closest papers by meaning
- Decoupling Representation and Classifier for Long-Tailed Recognition (ICLR 2020 · similarity 0.92)
- Improving Calibration for Long-Tailed Recognition (CVPR 2021 · similarity 0.91)
- No One Left Behind: Improving the Worst Categories in Long-Tailed Learning (CVPR 2023 · similarity 0.90)
- Long-tailed Recognition with Model Rebalancing (NeurIPS 2025 · similarity 0.89)
- Decoupled Optimisation for Long-Tailed Visual Recognition (AAAI 2024 · similarity 0.88)
- Exploring Weight Balancing on Long-Tailed Recognition Problem (ICLR 2024 · similarity 0.88)
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.
