mlsanityData to 7 Oct 2026

Shedding Light on Large Generative Networks: Estimating Epistemic Uncertainty in Diffusion Models

arXiv 2406.18580 ↗First posted 5 Jun 2024cs.CV cs.AI cs.LGcited by 22

On the map: Diffusion models → Diffusion foundations and sampling

Abstract (arXiv)

Generative diffusion models, notable for their large parameter count (exceeding 100 million) and operation within high-dimensional image spaces, pose significant challenges for traditional uncertainty estimation methods due to computational demands. In this work, we introduce an innovative framework, Diffusion Ensembles for Capturing Uncertainty (DECU), designed for estimating epistemic uncertainty for diffusion models. The DECU framework introduces a novel method that efficiently trains ensembles of conditional diffusion models by incorporating a static set of pre-trained parameters, drastically reducing the computational burden and the number of parameters that require training. Additionally, DECU employs Pairwise-Distance Estimators (PaiDEs) to accurately measure epistemic uncertainty by evaluating the mutual information between model outputs and weights in high-dimensional spaces. The effectiveness of this framework is demonstrated through experiments on the ImageNet dataset, highlighting its capability to capture epistemic uncertainty, specifically in under-sampled image classes.

Key facts. Cited by 22 papers in the map (data through 7 Oct 2026). Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models. Builds on 23 of its 43 references that are in the map. First posted on arXiv 5 Jun 2024.

Builds on

23 of 43 references.

Closest ancestors · 4

Same direction, newest first

  1. 2021High-Resolution Image Synthesis with Latent Diffusion Models20 Dec
  2. 2021Diffusion Models Beat GANs on Image Synthesis11 May
  3. 2020Denoising Diffusion Implicit Models6 Oct
  4. 2020Denoising Diffusion Probabilistic Models19 Jun

Other references · 19

  1. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
  2. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  3. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  4. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,322
  5. 2016XGBoost: A Scalable Tree Boosting Systemcited by 5,120
  6. 2016Simple and Scalable Predictive Uncertainty Estimation using Deep Ensemblescited by 3,888
  7. 2018Large Scale GAN Training for High Fidelity Natural Image Synthesiscited by 2,609
  8. 2017What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?cited by 2,335
  9. 2015Variational Inference with Normalizing Flowscited by 2,161
  10. 2019Aleatoric and Epistemic Uncertainty in Machine Learning: An Introduction to Concepts and Methodscited by 903
  11. 2017Deep Bayesian Active Learning with Image Datacited by 812
  12. 2018Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Modelscited by 748
7 more
  1. 2023GAIA-1: A Generative World Model for Autonomous Drivingcited by 618
  2. 2019BatchBALD: Efficient and Diverse Batch Acquisition for Deep Bayesian Active Learningcited by 428
  3. 2020Posterior Network: Uncertainty Estimation without OOD Samples via Density-Based Pseudo-Countscited by 163
  4. 2017Estimating Mixture Entropy with Pairwise Distancescited by 34
  5. 2017Synthetic Medical Images from Dual Generative Adversarial Networkscited by 32
  6. 2021Natural Posterior Network: Deep Bayesian Uncertainty for Exponential Family Distributionscited by 13
  7. 2023Normalizing Flow Ensembles for Rich Aleatoric and Epistemic Uncertainty Modelingcited by 10

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 2; 2025: 7; 2026: 13.

22 papers.

Sort:

By area:

  1. 2024Around the World in 80 Timesteps: A Generative Approach to Global Visual Geolocationcited by 23
  2. 2025Generative Uncertainty in Diffusion Modelscited by 20
  3. 2025World Models That Know When They Don't Know - Controllable Video Generation with Calibrated Uncertaintycited by 16
  4. 2024Towards Understanding and Quantifying Uncertainty for Text-to-Image Generationcited by 11
  5. 2025How Confident are Video Models? Empowering Video Models to Express their Uncertaintycited by 6
  6. 2026DiffusionRollout: Uncertainty-Aware Rollout Planning in Long-Horizon PDE Solvingcited by 4
  7. 2026Uncertainty Quantification for Flow-Based Vision-Language-Action Modelscited by 4
  8. 2025Trustworthy image-to-image translation: evaluating uncertainty calibration in unpaired training scenarioscited by 3
  9. 2026Flow Matching with Uncertainty Quantification and Guidancecited by 3
  10. 2025Improved probabilistic regression using diffusion modelscited by 2
  11. 2026Quantifying Epistemic Uncertainty in Diffusion Modelscited by 2
  12. 2025Perfecting Depth: Uncertainty-Aware Enhancement of Metric Depthcited by 1
10 more
  1. 2026Uncertainty-Aware Distribution-to-Distribution Flow Matching for Scientific Imagingcited by 1
  2. 2025EMoE: Training-Free Expert Disagreement for Uncertainty-Aware Text-to-Image Diffusioncited by 0
  3. 2026Efficient Credal Prediction through Decalibrationcited by 0
  4. 2026Hard to See, Hard to Label: Generative and Symbolic Acquisition for Subtle Visual Phenomenacited by 0
  5. 2026Uncertainty Quantification for Large Language Diffusion Modelscited by 0
  6. 2026Uncertainty Estimation for Molecular Diffusion Modelscited by 0
  7. 2026LM-X: Explainable Vision--Language--Action Modeling via Progress, Event, and Uncertainty Predictioncited by 0
  8. 2026Is your uncertainty map wrong, or is its target? Exact diagnostics for the Tweedie diagonal, and a gradient-free alternativecited by 0
  9. 2026Generative Uncertainty as a Self-supervised Signal for Semantic Similarity Learningcited by 0
  10. 2026Matisse: Evidence-Space Reasoning for Active 3D Reconstructioncited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Diffusion Epistemic Uncertainty with Asymmetric Learning for Diffusion-Generated Image Detection
  2. 2024Estimating Epistemic and Aleatoric Uncertainty with a Single Model
  3. 2024Diffusion Model Guided Sampling with Pixel-Wise Aleatoric Uncertainty Estimation
  4. 2024Zero-Shot Uncertainty Quantification using Diffusion Probabilistic Models
  5. 2023AdaDiff: Accelerating Diffusion Models through Step-Wise Adaptive Computation
  6. 2025Diffusion Models: A Mathematical Introduction
  7. 2023User-defined Event Sampling and Uncertainty Quantification in Diffusion Models for Physical Dynamical Systems
  8. 2025Generative Modeling with Bayesian Sample Inference