mlsanityData to 7 Oct 2026

Latent-based Diffusion Model for Long-tailed Recognition

arXiv 2404.04517 ↗First posted 6 Apr 2024cs.CV cs.AIcited by 14

On the map: Diffusion models → Diffusion foundations and sampling

Abstract (arXiv)

Long-tailed imbalance distribution is a common issue in practical computer vision applications. Previous works proposed methods to address this problem, which can be categorized into several classes: re-sampling, re-weighting, transfer learning, and feature augmentation. In recent years, diffusion models have shown an impressive generation ability in many sub-problems of deep computer vision. However, its powerful generation has not been explored in long-tailed problems. We propose a new approach, the Latent-based Diffusion Model for Long-tailed Recognition (LDMLR), as a feature augmentation method to tackle the issue. First, we encode the imbalanced dataset into features using the baseline model. Then, we train a Denoising Diffusion Implicit Model (DDIM) using these encoded features to generate pseudo-features. Finally, we train the classifier using the encoded and pseudo-features from the previous two steps. The model's accuracy shows an improvement on the CIFAR-LT and ImageNet-LT datasets by using the proposed method.

Key facts. Cited by 14 papers in the map (data through 7 Oct 2026). Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models. Builds on 28 of its 49 references that are in the map. First posted on arXiv 6 Apr 2024.

Builds on

28 of 49 references.

Closest ancestors · 5

Same direction, newest first

  1. 2023Class-Balancing Diffusion Models30 Apr
  2. 2022Scalable Diffusion Models with Transformers19 Dec
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Models20 Dec
  4. 2021Improved Denoising Diffusion Probabilistic Models18 Feb
  5. 2020Denoising Diffusion Implicit Models6 Oct

Other references · 23

  1. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  2. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,539
  3. 2019Learning Imbalanced Datasets with Label-Distribution-Aware Margin Losscited by 987
  4. 2019Decoupling Representation and Classifier for Long-Tailed Recognitioncited by 806
  5. 2017In Defense of the Triplet Loss for Person Re-Identificationcited by 784
  6. 2019Large-Scale Long-Tailed Recognition in an Open Worldcited by 704
  7. 2020Long-tail learning via logit adjustmentcited by 605
  8. 2020Balanced Meta-Softmax for Long-Tailed Visual Recognitioncited by 397
  9. 2021Distribution Alignment: A Unified Framework for Long-tail Visual Recognitioncited by 198
  10. 2020Disentangling Label Distribution for Long-tailed Visual Recognitioncited by 172
  11. 2020M2m: Imbalanced Classification via Major-to-minor Translationcited by 134
  12. 2016Range Loss for Deep Face Recognition with Long-tailcited by 111
11 more
  1. 2021The Majority Can Help The Minority: Context-rich Minority Oversampling for Long-tailed Classificationcited by 100
  2. 2021FASA: Feature Augmentation and Sampling Adaptation for Long-Tailed Instance Segmentationcited by 76
  3. 2021Label-Imbalanced and Group-Sensitive Classification under Overparameterizationcited by 74
  4. 2021Self Supervision to Distillation for Long-Tailed Visual Recognitioncited by 73
  5. 2016Factors in Finetuning Deep Model for object detectioncited by 53
  6. 2020Posterior Re-calibration for Imbalanced Datasetscited by 53
  7. 2022Long-Tailed Classification of Thorax Diseases on Chest X-Ray: A New Benchmark Studycited by 45
  8. 2022ProCo: Prototype-aware Contrastive Learning for Long-tailed Medical Image Classificationcited by 25
  9. 2022Do Deep Networks Transfer Invariances Across Classes?cited by 12
  10. 2023Transfer Knowledge from Head to Tail: Uncertainty Calibration under Long-tailed Distributioncited by 12
  11. 2023Wrapped Cauchy Distributed Angular Softmax for Long-Tailed Visual Recognitioncited by 11

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 3; 2025: 5; 2026: 6.

14 papers.

Sort:

By area:

  1. 2025A Comprehensive Survey on Imbalanced Data Learningcited by 10
  2. 2024Diffusion Curriculum: Synthetic-to-Real Data Curriculum via Image-Guided Diffusioncited by 6
  3. 2024Improving Viewpoint Consistency in 3D Generation via Structure Feature and CLIP Guidancecited by 2
  4. 2025Enhancing Features in Long-tailed Data Using Large Vision Modelcited by 2
  5. 2025Synthetic Data Augmentation using Pre-trained Diffusion Models for Long-tailed Food Image Classificationcited by 2
  6. 2025Language-Informed Hyperspectral Image Synthesis for Imbalanced-Small Sample Classification via Semi-Supervised Conditional Diffusion Modelcited by 1
  7. 2025Structural Energy-Guided Sampling for View-Consistent Text-to-3Dcited by 1
  8. 2026Decision Boundary-aware Generation for Long-tailed Learningcited by 1
  9. 2024Anisotropic Diffusion Probabilistic Model for Imbalanced Image Classificationcited by 0
  10. 2026Disentangling Hardness from Noise: An Uncertainty-Driven Model-Agnostic Framework for Long-Tailed Remote Sensing Classificationcited by 0
  11. 2026Semantics-Aware Generative Latent Data Augmentation for Learning in Low-Resource Domainscited by 0
  12. 2026Structural Energy Guidance for View-Consistent Text-to-3D Generationcited by 0
2 more
  1. 2026Label Shift Aware Adaptation for Online Zero-shot Learning with Contrastive Language-Image Pre-Training (CLIP)cited by 0
  2. 2026DiffReID: Discriminative Diffusion Model for Object Re-Identificationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024DiffuLT: How to Make Diffusion Model Useful for Long-tail Recognition
  2. 2022Diffusion Models in Vision: A Survey
  3. 2024Diffusion Models and Representation Learning: A Survey
  4. 2024Training Class-Imbalanced Diffusion Model Via Overlap Optimization
  5. 2024Long Tail Image Generation Through Feature Space Augmentation and Iterated Learning
  6. 2025DDB: Diffusion Driven Balancing to Address Spurious Correlations
  7. 2026LDM-is-AE: Latent Diffusion Model is an Auto-Encoder for End-to-End Image Generation
  8. 2025CORAL: Disentangling Latent Representations in Long-Tailed Diffusion