Research map / Label-efficient and robust learning / Semi-supervised learning
Learning Dynamics of Logits Debiasing for Long-Tailed Semi-Supervised Learning
Yue Cheng, Jiajun Zhang, Xiaohui Gao, Weiwei Xing, Zhanxing Zhu
Read the paperCheck your idea against 137,217 papers
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 |
|---|---|---|---|
| add DyTrim on top of CDMAD | FixMatch+CDMAD · CIFAR-10-LT (gamma=100) | bACC: 83.6 -> 84.8 | Improved |
| add DyTrim on top of CDMAD | FixMatch+CDMAD · CIFAR-10-LT (gamma=150) | bACC: 80.8 -> 82.0 | Improved |
| add DyTrim on top of CDMAD | FixMatch+CDMAD · CIFAR-10-LT (gamma_l=100, gamma_u=1) | bACC: 87.5 -> 88.9 | Improved |
| add DyTrim on top of CCL | FixMatch+CCL · CIFAR-10-LT (gamma=100) | bACC: 86.2 -> 86.7 | Improved |
| add DyTrim on top of CCL | FixMatch+CCL · CIFAR-10-LT (gamma=150) | bACC: 84.0 -> 84.0 | No clear effect |
| add DyTrim on top of CDMAD | FixMatch+CDMAD · STL-10-LT (gamma_l=20) | bACC: 75.2 -> 77.9 | Improved |
| add DyTrim on top of CCL | FixMatch+CCL · STL-10-LT (gamma_l=20) | bACC: 83.1 -> 83.3 | No clear effect |
| replace CDMAD with DyTrim | FlexMatch · CIFAR-10-LT (gamma=100) | bACC: 68.4 -> 70.7 | Improved |
| replace CDMAD with DyTrim | FreeMatch · CIFAR-10-LT (gamma=100) | bACC: 69.9 -> 72.3 | Improved |
| replace CDMAD with DyTrim | FixMatch · CIFAR-100-LT (gamma=100) | bACC: 44.1 -> 44.8 | Improved |
| add DyTrim to FixMatch instead of CDMAD | FixMatch · ImageNet-LT | bACC: 35.4 -> 37.2 | Improved |
| replace CDMAD with DyTrim (TinyViT) | FixMatch TinyViT · CIFAR-10-LT (gamma_u=150) | bACC: 45.4 -> 47.3 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 4 of 4 on most shared results.
- FixMatch — FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence (NeurIPS 2020) · better on 7 of 7 shared results
- CoSSL — CoSSL: Co-Learning of Representation and Classifier for Imbalanced Semi-Supervised Learning (CVPR 2022) · better on 4 of 4 shared results
- FlexMatch* — FlexMatch: Boosting Semi-Supervised Learning with Curriculum Pseudo Labeling (NeurIPS 2021) · better on 2 of 2 shared results
- FreeMatch* — FreeMatch: Self-adaptive Thresholding for Semi-supervised Learning (ICLR 2023) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Continuous Contrastive Learning for Long-Tailed Semi-Supervised Recognition (NeurIPS 2024 · similarity 0.88)
- BEM: Balanced and Entropy-based Mix for Long-Tailed Semi-Supervised Learning (CVPR 2024 · similarity 0.87)
- A Square Peg in a Square Hole: Meta-Expert for Long-Tailed Semi-Supervised Learning (ICML 2025 · similarity 0.87)
- CDMAD: Class-Distribution-Mismatch-Aware Debiasing for Class-Imbalanced Semi-Supervised Learning (CVPR 2024 · similarity 0.87)
- Erasing the Bias: Fine-Tuning Foundation Models for Semi-Supervised Learning (ICML 2024 · similarity 0.87)
- Keep It on a Leash: Controllable Pseudo-label Generation Towards Realistic Long-Tailed Semi-Supervised Learning (NeurIPS 2025 · similarity 0.87)
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.
