mlsanityData to 7 Oct 2026

How I Met Your Bias: Investigating Bias Amplification in Diffusion Models

arXiv 2512.20233 ↗First posted 23 Dec 2025cs.LG cs.CVcited by 4

On the map: Diffusion models → Diffusion foundations and sampling

Abstract (arXiv)

Diffusion-based generative models demonstrate state-of-the-art performance across various image synthesis tasks, yet their tendency to replicate and amplify dataset biases remains poorly understood. Although previous research has viewed bias amplification as an inherent characteristic of diffusion models, this work provides the first analysis of how sampling algorithms and their hyperparameters influence bias amplification. We empirically demonstrate that samplers for diffusion models -- commonly optimized for sample quality and speed -- have a significant and measurable effect on bias amplification. Through controlled studies with models trained on Biased MNIST, Multi-Color MNIST and BFFHQ, and with Stable Diffusion, we show that sampling hyperparameters can induce both bias reduction and amplification, even when the trained model is fixed. Source code is available at https://github.com/How-I-met-your-bias/how_i_met_your_bias.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models. Builds on 43 of its 47 references that are in the map. First posted on arXiv 23 Dec 2025.

Builds on

43 of 47 references.

Closest ancestors · 15

Same direction, newest first

  1. 2025Frequency Regulation for Exposure Bias Mitigation in Diffusion Models14 Jul
  2. 2025Deeper Diffusion Models Amplify Bias23 May
  3. 2025Diffusing DeBias: Synthetic Bias Amplification for Model Debiasing13 Feb
  4. 2023Unmasking Bias in Diffusion Model Training12 Oct
  5. 2023Elucidating the Exposure Bias in Diffusion Models29 Aug
  6. 2023Alleviating Exposure Bias in Diffusion Models through Sampling with Shifted Time Steps24 May
  7. 2023Input Perturbation Reduces Exposure Bias in Diffusion Models27 Jan
  8. 2022Classifier-Free Diffusion Guidance26 Jul
  9. 2022DPM-Solver: A Fast ODE Solver for Diffusion Probabilistic Model Sampling in Around 10 Steps2 Jun
  10. 2022Elucidating the Design Space of Diffusion-Based Generative Models1 Jun
  11. 2021High-Resolution Image Synthesis with Latent Diffusion Models20 Dec
  12. 2021Diffusion Models Beat GANs on Image Synthesis11 May
3 more
  1. 2020Score-Based Generative Modeling through Stochastic Differential Equations26 Nov
  2. 2020Denoising Diffusion Implicit Models6 Oct
  3. 2020Denoising Diffusion Probabilistic Models19 Jun

Other references · 28

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  3. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 5,986
  4. 2023Improved Baselines with Visual Instruction Tuningcited by 4,362
  5. 2021Zero-Shot Text-to-Image Generationcited by 4,276
  6. 2022LAION-5B: An open large-scale dataset for training next generation image-text modelscited by 4,131
  7. 2023Human Preference Score v2: A Solid Benchmark for Evaluating Human Preferences of Text-to-Image Synthesiscited by 847
  8. 2022Easily Accessible Text-to-Image Generation Amplifies Demographic Stereotypes at Large Scalecited by 262
  9. 2017GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibriumcited by 208
  10. 2019Learning De-biased Representations with Biased Representationscited by 199
  11. 2021Learning Debiased Representation via Disentangled Feature Augmentationcited by 116
  12. 2021BiaSwap: Removing dataset bias with bias-tailored swapping augmentationcited by 70
16 more
  1. 2023The Bias Amplification Paradox in Text-to-Image Generationcited by 69
  2. 2024OpenBias: Open-set Bias Detection in Text-to-Image Generative Modelscited by 61
  3. 2023Discovering and Mitigating Visual Biases through Keyword Explanationcited by 58
  4. 2023Uncurated Image-Text Datasets: Shedding Light on Demographic Biascited by 51
  5. 2024Balancing Act: Distribution-Guided Debiasing in Diffusion Modelscited by 50
  6. 2022Discover and Mitigate Unknown Biases with Debiasing Alternate Networkscited by 47
  7. 2023T2IAT: Measuring Valence and Stereotypical Biases in Text-to-Image Generationcited by 39
  8. 2023Debiasing Scores and Prompts of 2D Diffusion for View-consistent Text-to-3D Generationcited by 32
  9. 2025Dissecting and Mitigating Diffusion Bias via Mechanistic Interpretabilitycited by 26
  10. 2023Quantifying Bias in Text-to-Image Generative Modelscited by 16
  11. 2023DREAM: Diffusion Rectification and Estimation-Adaptive Modelscited by 12
  12. 2024Debiasing Text-to-Image Diffusion Modelscited by 10
  13. 2025Responsible Diffusion Models via Constraining Text Embeddings within Safe Regionscited by 8
  14. 2024Generated Bias: Auditing Internal Bias Dynamics of Text-To-Image Generative Modelscited by 3
  15. 2024Gender Bias Evaluation in Text-to-image Generation: A Surveycited by 2
  16. 2024Bias Begets Bias: The Impact of Biased Embeddings on Diffusion Modelscited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 4.

4 papers.

  1. 2026Stay Fair! Ensuring Group Fairness in Diffusion Models Across Guidance Scalescited by 2
  2. 2026Position: Don't Just "Fix it in Post": A Science of AI Must Study Training Dynamicscited by 2
  3. 2026Diffusion models recover accurate mixture weights despite score function insensitivitycited by 2
  4. 2026Efficient bias mitigation in T2I diffusion models using Concept Graphscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Shaping Inductive Bias in Diffusion Models through Frequency-Based Noise Control
  2. 2024Advancing Diffusion Models: Alias-Free Resampling and Enhanced Rotational Equivariance
  3. 2024Would Deep Generative Models Amplify Bias in Future Models?
  4. 2022Learning Fast Samplers for Diffusion Models by Differentiating Through Sample Quality
  5. 2022Progressive Distillation for Fast Sampling of Diffusion Models
  6. 2026Biased Generalization in Diffusion Models
  7. 2023Exploiting the Signal-Leak Bias in Diffusion Models
  8. 2023Mitigating Exposure Bias in Discriminator Guided Diffusion Models