# Balancing Logit Variation for Long-tailed Semantic Segmentation

arXiv: https://arxiv.org/abs/2306.02061 · HTML: https://mlsanity.com/paper/2306.02061/

## Key facts

- Cited by 26 papers in the map (data through 14 Oct 2026).
- Builds on 100 of its 138 references that are in the map.
- First posted on arXiv 3 Jun 2023.

## Builds on (most cited first, 30 of 100)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [Fully Convolutional Networks for Semantic Segmentation](https://mlsanity.com/paper/1411.4038/) (2014)
- [The Cityscapes Dataset for Semantic Urban Scene Understanding](https://mlsanity.com/paper/1604.01685/) (2016)
- [DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs](https://mlsanity.com/paper/1606.00915/) (2016)
- [SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers](https://mlsanity.com/paper/2105.15203/) (2021)
- [CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features](https://mlsanity.com/paper/1905.04899/) (2019)
- [Pyramid Scene Parsing Network](https://mlsanity.com/paper/1612.01105/) (2016)
- [FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence](https://mlsanity.com/paper/2001.07685/) (2020)
- [Deep High-Resolution Representation Learning for Visual Recognition](https://mlsanity.com/paper/1908.07919/) (2019)
- [Rethinking Semantic Segmentation from a Sequence-to-Sequence Perspective with Transformers](https://mlsanity.com/paper/2012.15840/) (2020)
- [Deep High-Resolution Representation Learning for Human Pose Estimation](https://mlsanity.com/paper/1902.09212/) (2019)
- [Learning Transferable Features with Deep Adaptation Networks](https://mlsanity.com/paper/1502.02791/) (2015)
- [Self-training with Noisy Student improves ImageNet classification](https://mlsanity.com/paper/1911.04252/) (2019)
- [Unified Perceptual Parsing for Scene Understanding](https://mlsanity.com/paper/1807.10221/) (2018)
- [Class-Balanced Loss Based on Effective Number of Samples](https://mlsanity.com/paper/1901.05555/) (2019)
- [LVIS: A Dataset for Large Vocabulary Instance Segmentation](https://mlsanity.com/paper/1908.03195/) (2019)
- [CyCADA: Cycle-Consistent Adversarial Domain Adaptation](https://mlsanity.com/paper/1711.03213/) (2017)
- [Dual Attention Network for Scene Segmentation](https://mlsanity.com/paper/1809.02983/) (2018)
- [Adversarial Machine Learning at Scale](https://mlsanity.com/paper/1611.01236/) (2016)
- [Playing for Data: Ground Truth from Computer Games](https://mlsanity.com/paper/1608.02192/) (2016)
- [CCNet: Criss-Cross Attention for Semantic Segmentation](https://mlsanity.com/paper/1811.11721/) (2018)
- [Learning to Reweight Examples for Robust Deep Learning](https://mlsanity.com/paper/1803.09050/) (2018)
- [ADVENT: Adversarial Entropy Minimization for Domain Adaptation in Semantic Segmentation](https://mlsanity.com/paper/1811.12833/) (2018)
- [Long-tail learning via logit adjustment](https://mlsanity.com/paper/2007.07314/) (2020)
- [Learning to Adapt Structured Output Space for Semantic Segmentation](https://mlsanity.com/paper/1802.10349/) (2018)
- [Segmentation Transformer: Object-Contextual Representations for Semantic Segmentation](https://mlsanity.com/paper/1909.11065/) (2019)
- [Vision Transformer Adapter for Dense Predictions](https://mlsanity.com/paper/2205.08534/) (2022)

## Built on by (most cited first, 26 of 26)

- [A Systematic Review on Long-Tailed Learning](https://mlsanity.com/paper/2408.00483/) (2024)
- [DropPos: Pre-Training Vision Transformers by Reconstructing Dropped Positions](https://mlsanity.com/paper/2309.03576/) (2023)
- [Delving into the Trajectory Long-tail Distribution for Muti-object Tracking](https://mlsanity.com/paper/2403.04700/) (2024)
- [AUCSeg: AUC-oriented Pixel-level Long-tail Semantic Segmentation](https://mlsanity.com/paper/2409.20398/) (2024)
- [Using Unreliable Pseudo-Labels for Label-Efficient Semantic Segmentation](https://mlsanity.com/paper/2306.02314/) (2023)
- [Instance-Warp: Saliency Guided Image Warping for Unsupervised Domain Adaptation](https://mlsanity.com/paper/2403.12712/) (2024)
- [Data Optimization in Deep Learning: A Survey](https://mlsanity.com/paper/2310.16499/) (2023)
- [Bootstrap Masked Visual Modeling via Hard Patches Mining](https://mlsanity.com/paper/2312.13714/) (2023)
- [Progressive Proxy Anchor Propagation for Unsupervised Semantic Segmentation](https://mlsanity.com/paper/2407.12463/) (2024)
- [Semantic Connectivity-Driven Pseudo-labeling for Cross-domain Segmentation](https://mlsanity.com/paper/2312.06331/) (2023)
- [Frequency-based Matcher for Long-tailed Semantic Segmentation](https://mlsanity.com/paper/2406.03917/) (2024)
- [Understanding Representation of Deep Equilibrium Models from Neural Collapse Perspective](https://mlsanity.com/paper/2410.23391/) (2024)
- [DUDA: Distilled Unsupervised Domain Adaptation for Lightweight Semantic Segmentation](https://mlsanity.com/paper/2504.09814/) (2025)
- [Adaptive Location Hierarchy Learning for Long-Tailed Mobility Prediction](https://mlsanity.com/paper/2505.19965/) (2025)
- [Keep It on a Leash: Controllable Pseudo-label Generation Towards Realistic Long-Tailed Semi-Supervised Learning](https://mlsanity.com/paper/2510.03993/) (2025)
- [Denoise and Align: Towards Source-Free UDA for Robust Panoramic Semantic Segmentation](https://mlsanity.com/paper/2603.25131/) (2026)
- [Why Do Vision Language Models Struggle To Recognize Human Emotions?](https://mlsanity.com/paper/2604.15280/) (2026)
- [PAT: Pixel-wise Adaptive Training for Long-tailed Segmentation](https://mlsanity.com/paper/2404.05393/) (2024)
- [Addressing Imbalance for Class Incremental Learning in Medical Image Classification](https://mlsanity.com/paper/2407.13768/) (2024)
- [Mitigating Data Imbalance in Automated Speaking Assessment](https://mlsanity.com/paper/2509.03010/) (2025)
- [DPSformer: A long-tail-aware model for improving heavy rainfall prediction](https://mlsanity.com/paper/2509.25208/) (2025)
- [NS-Pep: De novo Peptide Design with Non-Standard Amino Acids](https://mlsanity.com/paper/2510.03326/) (2025)
- [MuFFIN: Multifaceted Pronunciation Feedback Model with Interactive Hierarchical Neural Modeling](https://mlsanity.com/paper/2510.04956/) (2025)
- [Classifier Enhancement Using Extended Context and Domain Experts for Semantic Segmentation](https://mlsanity.com/paper/2510.25174/) (2025)
- [SOFTooth: Semantics-Enhanced Order-Aware Fusion for Tooth Instance Segmentation](https://mlsanity.com/paper/2512.23411/) (2025)
- [SAM3 Self-Distillation for Fine-Grained GOOSE 2D Semantic Segmentation](https://mlsanity.com/paper/2606.20130/) (2026)

## Abstract (arXiv)

Semantic segmentation usually suffers from a long-tail data distribution. Due to the imbalanced number of samples across categories, the features of those tail classes may get squeezed into a narrow area in the feature space. Towards a balanced feature distribution, we introduce category-wise variation into the network predictions in the training phase such that an instance is no longer projected to a feature point, but a small region instead. Such a perturbation is highly dependent on the category scale, which appears as assigning smaller variation to head classes and larger variation to tail classes. In this way, we manage to close the gap between the feature areas of different categories, resulting in a more balanced representation. It is noteworthy that the introduced variation is discarded at the inference stage to facilitate a confident prediction. Although with an embarrassingly simple implementation, our method manifests itself in strong generalizability to various datasets and task settings. Extensive experiments suggest that our plug-in design lends itself well to a range of state-of-the-art approaches and boosts the performance on top of them.