mlsanityData to 7 Oct 2026

Provable Sample-Efficient Transfer Learning Conditional Diffusion Models via Representation Learning

arXiv 2502.04491 ↗First posted 6 Feb 2025cs.LG math.ST stat.MLcited by 5

On the map: Diffusion models → Diffusion foundations and sampling

Abstract (arXiv)

While conditional diffusion models have achieved remarkable success in various applications, they require abundant data to train from scratch, which is often infeasible in practice. To address this issue, transfer learning has emerged as an essential paradigm in small data regimes. Despite its empirical success, the theoretical underpinnings of transfer learning conditional diffusion models remain unexplored. In this paper, we take the first step towards understanding the sample efficiency of transfer learning conditional diffusion models through the lens of representation learning. Inspired by practical training procedures, we assume that there exists a low-dimensional representation of conditions shared across all tasks. Our analysis shows that with a well-learned representation from source tasks, the samplecomplexity of target tasks can be reduced substantially. In addition, we investigate the practical implications of our theoretical results in several real-world applications of conditional diffusion models. Numerical experiments are also conducted to verify our results.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models. Builds on 51 of its 65 references that are in the map. First posted on arXiv 6 Feb 2025.

Builds on

51 of 65 references.

Closest ancestors · 18

Same direction, newest first

  1. 2024On Statistical Rates of Conditional Diffusion Transformers: Approximation, Estimation and Minimax Optimality26 Nov
  2. 2024Model Free Prediction with Uncertainty Assessment21 May
  3. 2024Unveil Conditional Diffusion Models with Classifier-free Guidance: A Sharp Statistical Theory18 Mar
  4. 2024Scaling Rectified Flow Transformers for High-Resolution Image Synthesis5 Mar
  5. 2023Hierarchical Semi-Implicit Variational Inference with Application to Diffusion Model Acceleration26 Oct
  6. 2023Diffusion with Forward Models: Solving Stochastic Inverse Problems Without Direct Supervision20 Jun
  7. 2023The probability flow ODE is provably fast19 May
  8. 2023DiffFit: Unlocking Transferability of Large Diffusion Models via Simple Parameter-Efficient Fine-Tuning13 Apr
  9. 2023Diffusion Models are Minimax Optimal Distribution Estimators3 Mar
  10. 2023Score Approximation, Estimation and Distribution Recovery of Diffusion Models on Low-Dimensional Data14 Feb
  11. 2022Improved Analysis of Score-based Generative Modeling: User-Friendly Bounds under Minimal Smoothness Assumptions3 Nov
  12. 2022Convergence of score-based generative modeling for general data distributions26 Sep
6 more
  1. 2022Sampling is as easy as learning the score: theory for diffusion models with minimal data assumptions22 Sep
  2. 2022Classifier-Free Diffusion Guidance26 Jul
  3. 2022Few-Shot Diffusion Models30 May
  4. 2021High-Resolution Image Synthesis with Latent Diffusion Models20 Dec
  5. 2020Score-Based Generative Modeling through Stochastic Differential Equations26 Nov
  6. 2020Denoising Diffusion Probabilistic Models19 Jun

Other references · 33

  1. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
  2. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,539
  3. 2013Auto-Encoding Variational Bayescited by 3,980
  4. 2022DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generationcited by 3,286
  5. 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusioncited by 2,730
  6. 2022Planning with Diffusion for Flexible Behavior Synthesiscited by 1,023
  7. 2021CSDI: Conditional Score-based Diffusion Models for Probabilistic Time Series Imputationcited by 577
  8. 2016Error bounds for approximations with deep ReLU networkscited by 567
  9. 2022Diffusion Policies as an Expressive Policy Class for Offline Reinforcement Learningcited by 543
  10. 2021Solving Inverse Problems in Medical Imaging with Score-Based Generative Modelscited by 514
  11. 2022Is Conditional Generative Modeling all you need for Decision-Making?cited by 504
  12. 2017Nonparametric regression using deep neural networks with ReLU activation functioncited by 431
21 more
  1. 2023AudioLDM 2: Learning Holistic Audio Generation with Self-supervised Pretrainingcited by 343
  2. 2023SVDiff: Compact Parameter Space for Diffusion Fine-Tuningcited by 338
  3. 2021Autoregressive Denoising Diffusion Models for Multivariate Probabilistic Time Series Forecastingcited by 332
  4. 2023Goal-Conditioned Imitation Learning using Score-based Diffusion Policiescited by 229
  5. 2020Few-Shot Learning via Learning the Representation, Provablycited by 213
  6. 2015The Benefit of Multitask Representation Learningcited by 194
  7. 2020On the Theory of Transfer Learning: The Importance of Task Diversitycited by 186
  8. 2020Provable Meta-Learning of Linear Representationscited by 161
  9. 2023Protein Design with Guided Discrete Diffusioncited by 154
  10. 2022Solving 3D Inverse Problems using Pre-trained 2D Diffusion Modelscited by 130
  11. 2023Diffusion Model is an Effective Planner and Data Synthesizer for Multi-Task Reinforcement Learningcited by 114
  12. 2021D2C: Diffusion-Denoising Models for Few-shot Conditional Generationcited by 96
  13. 2019Nonparametric Regression on Low-Dimensional Manifolds using Deep ReLU Networks : Function Approximation and Statistical Recoverycited by 73
  14. 2023MetaDiffuser: Diffusion Model as Conditional Planner for Offline Meta-RLcited by 65
  15. 2024Optimal score estimation via empirical Bayes smoothingcited by 60
  16. 2021Offline Meta-Reinforcement Learning with Online Self-Supervisioncited by 52
  17. 2023Reward-Directed Conditional Diffusion: Provable Distribution Estimation and Reward Improvementcited by 49
  18. 2021How Fine-Tuning Allows for Effective Meta-Learningcited by 41
  19. 2014Dimension-independent likelihood-informed MCMCcited by 36
  20. 2020Distribution Approximation and Statistical Estimation Guarantees of Generative Adversarial Networkscited by 23
  21. 2023Metalearning with Very Few Samples Per Taskcited by 11

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 1; 2026: 4.

5 papers.

  1. 2026On the Limits of Latent Reuse in Diffusion Modelscited by 1
  2. 2025GUST: Quantifying Free-Form Geometric Uncertainty of Metamaterials Using Small Datacited by 0
  3. 2026Sample Complexity of Transfer Learning: An Optimal Transport Approachcited by 0
  4. 2026Multi-Objective Learning for Diffusion Models: A Statistical Theory under Semi-Supervised Learningcited by 0
  5. 2026Robustness of Diffusion Models under Distribution Shiftcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Transfer Learning for Diffusion Models
  2. 2024An Overview of Diffusion Models: Applications, Guided Generation, Statistical Rates and Optimization
  3. 2025Learning Diffusion Models with Flexible Representation Guidance
  4. 2023Bridging the Gap: Addressing Discrepancies in Diffusion Model Training for Classifier-Free Guidance
  5. 2023The Emergence of Reproducibility and Generalizability in Diffusion Models
  6. 2023Diffusion Model as Representation Learner
  7. 2026Understanding diffusion models requires rethinking (again) generalization
  8. 2024Conditional sampling within generative diffusion models