# Denoising Task Difficulty-based Curriculum for Training Diffusion Models

arXiv: https://arxiv.org/abs/2403.10348 · HTML: https://mlsanity.com/paper/2403.10348/

## Key facts

- Cited by 27 papers in the map (data through 14 Oct 2026).
- Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models.
- Builds on 58 of its 76 references that are in the map.
- First posted on arXiv 15 Mar 2024.

## Placed in

- Diffusion models / [Diffusion foundations and sampling](https://mlsanity.com/diffusion/diffusion-foundations-and-sampling/)

## Builds on (most cited first, 30 of 58)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [Score-Based Generative Modeling through Stochastic Differential Equations](https://mlsanity.com/paper/2011.13456/) (2020)
- [Diffusion Models Beat GANs on Image Synthesis](https://mlsanity.com/paper/2105.05233/) (2021)
- [Deep Unsupervised Learning using Nonequilibrium Thermodynamics](https://mlsanity.com/paper/1503.03585/) (2015)
- [Hierarchical Text-Conditional Image Generation with CLIP Latents](https://mlsanity.com/paper/2204.06125/) (2022)
- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [A Style-Based Generator Architecture for Generative Adversarial Networks](https://mlsanity.com/paper/1812.04948/) (2018)
- [Classifier-Free Diffusion Guidance](https://mlsanity.com/paper/2207.12598/) (2022)
- [Generative Modeling by Estimating Gradients of the Data Distribution](https://mlsanity.com/paper/1907.05600/) (2019)
- [Improved Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2102.09672/) (2021)
- [Elucidating the Design Space of Diffusion-Based Generative Models](https://mlsanity.com/paper/2206.00364/) (2022)
- [Improved Techniques for Training GANs](https://mlsanity.com/paper/1606.03498/) (2016)
- [DPM-Solver: A Fast ODE Solver for Diffusion Probabilistic Model Sampling in Around 10 Steps](https://mlsanity.com/paper/2206.00927/) (2022)
- [Consistency Models](https://mlsanity.com/paper/2303.01469/) (2023)
- [Imagen Video: High Definition Video Generation with Diffusion Models](https://mlsanity.com/paper/2210.02303/) (2022)
- [DiffWave: A Versatile Diffusion Model for Audio Synthesis](https://mlsanity.com/paper/2009.09761/) (2020)
- [Cascaded Diffusion Models for High Fidelity Image Generation](https://mlsanity.com/paper/2106.15282/) (2021)
- [Improved Precision and Recall Metric for Assessing Generative Models](https://mlsanity.com/paper/1904.06991/) (2019)
- [eDiff-I: Text-to-Image Diffusion Models with an Ensemble of Expert Denoisers](https://mlsanity.com/paper/2211.01324/) (2022)
- [SiT: Exploring Flow and Diffusion-based Generative Models with Scalable Interpolant Transformers](https://mlsanity.com/paper/2401.08740/) (2024)
- [SyncDreamer: Generating Multiview-consistent Images from a Single-view Image](https://mlsanity.com/paper/2309.03453/) (2023)
- [Improved Techniques for Training Consistency Models](https://mlsanity.com/paper/2310.14189/) (2023)
- [Analyzing and Improving the Training Dynamics of Diffusion Models](https://mlsanity.com/paper/2312.02696/) (2023)
- [Understanding Diffusion Objectives as the ELBO with Simple Data Augmentation](https://mlsanity.com/paper/2303.00848/) (2023)

## Built on by (most cited first, 27 of 27)

- [Curriculum Direct Preference Optimization for Diffusion and Consistency Models](https://mlsanity.com/paper/2405.13637/) (2024)
- [A Closer Look at Time Steps is Worthy of Triple Speed-Up for Diffusion Model Training](https://mlsanity.com/paper/2405.17403/) (2024)
- [Text-to-3D by Stitching a Multi-view Reconstruction Network to a Video Generator](https://mlsanity.com/paper/2510.13454/) (2025)
- [Decouple-Then-Merge: Finetune Diffusion Models as Multi-Task Learning](https://mlsanity.com/paper/2410.06664/) (2024)
- [UltraFlux: Data-Model Co-Design for High-quality Native 4K Text-to-Image Generation across Diverse Aspect Ratios](https://mlsanity.com/paper/2511.18050/) (2025)
- [Improved Immiscible Diffusion: Accelerate Diffusion Training by Reducing Its Miscibility](https://mlsanity.com/paper/2505.18521/) (2025)
- [Provable Benefit of Curriculum in Transformer Tree-Reasoning Post-Training](https://mlsanity.com/paper/2511.07372/) (2025)
- [Understanding, Accelerating, and Improving MeanFlow Training](https://mlsanity.com/paper/2511.19065/) (2025)
- [Beyond the Golden Data: Resolving the Motion-Vision Quality Dilemma via Timestep Selective Training](https://mlsanity.com/paper/2603.25527/) (2026)
- [Expert-Choice Routing Enables Adaptive Computation in Diffusion Language Models](https://mlsanity.com/paper/2604.01622/) (2026)
- [FEB-Cache: Frequency-Guided Exposure Bias Reduction for Enhancing Diffusion Transformer Caching](https://mlsanity.com/paper/2503.07120/) (2025)
- [From Fragments to Facts: A Curriculum-Driven DPO Approach for Generating Hindi News Veracity Explanations](https://mlsanity.com/paper/2507.05179/) (2025)
- [RDTF: Resource-efficient Dual-mask Training Framework for Multi-frame Animated Sticker Generation](https://mlsanity.com/paper/2503.17735/) (2025)
- [Adaptively Point-weighting Curriculum Learning](https://mlsanity.com/paper/2505.01665/) (2025)
- [NoiseRater: Meta-Learned Noise Valuation for Diffusion Model Training](https://mlsanity.com/paper/2605.08144/) (2026)
- [One-Sided Quantile Coupling for Flow Matching](https://mlsanity.com/paper/2608.00978/) (2026)
- [BrainTaskonomy: Learning How to Pretrain and What to Transfer in fMRI Foundation Models](https://mlsanity.com/paper/2609.10518/) (2026)
- [Curriculum-DPO++: Direct Preference Optimization via Data and Model Curricula for Text-to-Image Generation](https://mlsanity.com/paper/2602.13055/) (2026)
- [Level Up: Defining and Exploiting Transitional Problems for Curriculum Learning](https://mlsanity.com/paper/2603.13761/) (2026)
- [Can We Change the Stroke Size for Easier Diffusion?](https://mlsanity.com/paper/2603.26783/) (2026)
- [Learning Hierarchical and Geometry-Aware Graph Representations for Text-to-CAD](https://mlsanity.com/paper/2604.10075/) (2026)
- [AniMatrix: An Anime Video Generation Model that Thinks in Art, Not Physics](https://mlsanity.com/paper/2605.03652/) (2026)
- [Composable multi-satellite precipitation estimation for evolving observing systems](https://mlsanity.com/paper/2605.14426/) (2026)
- [Curriculum-guided Change Detection Training: Toward Accurate Serac Fall Monitoring](https://mlsanity.com/paper/2606.28012/) (2026)
- [CenSynCMB: Centre Maps and Physics-Guided Synthesis for Microbleed Detection](https://mlsanity.com/paper/2607.05325/) (2026)
- [Timestep Weighting: A Hidden Key to Effective ELBO-Based Flow-Matching RL](https://mlsanity.com/paper/2609.32665/) (2026)
- [NV-Tesseract-AD: Curriculum-Enhanced Conditional Diffusion for Multivariate Time-Series Anomaly Detection](https://mlsanity.com/paper/2610.12754/) (2026)

## Abstract (arXiv)

Diffusion-based generative models have emerged as powerful tools in the realm of generative modeling. Despite extensive research on denoising across various timesteps and noise levels, a conflict persists regarding the relative difficulties of the denoising tasks. While various studies argue that lower timesteps present more challenging tasks, others contend that higher timesteps are more difficult. To address this conflict, our study undertakes a comprehensive examination of task difficulties, focusing on convergence behavior and changes in relative entropy between consecutive probability distributions across timesteps. Our observational study reveals that denoising at earlier timesteps poses challenges characterized by slower convergence and higher relative entropy, indicating increased task difficulty at these lower timesteps. Building on these observations, we introduce an easy-to-hard learning scheme, drawing from curriculum learning, to enhance the training process of diffusion models. By organizing timesteps or noise levels into clusters and training models with ascending orders of difficulty, we facilitate an order-aware training regime, progressing from easier to harder denoising tasks, thereby deviating from the conventional approach of training diffusion models simultaneously across all timesteps. Our approach leads to improved performance and faster convergence by leveraging benefits of curriculum learning, while maintaining orthogonality with existing improvements in diffusion training techniques. We validate these advantages through comprehensive experiments in image generation tasks, including unconditional, class-conditional, and text-to-image generation.