mlsanityData to 13 Oct 2026

Optimizing for the Shortest Path in Denoising Diffusion Model

arXiv 2503.03265 ↗First posted 5 Mar 2025cs.CVcited by 6

On the map: Diffusion models → Diffusion foundations and sampling

Abstract (arXiv)

In this research, we propose a novel denoising diffusion model based on shortest-path modeling that optimizes residual propagation to enhance both denoising efficiency and quality. Drawing on Denoising Diffusion Implicit Models (DDIM) and insights from graph theory, our model, termed the Shortest Path Diffusion Model (ShortDF), treats the denoising process as a shortest-path problem aimed at minimizing reconstruction error. By optimizing the initial residuals, we improve the efficiency of the reverse diffusion process and the quality of the generated samples. Extensive experiments on multiple standard benchmarks demonstrate that ShortDF significantly reduces diffusion time (or steps) while enhancing the visual fidelity of generated samples compared to prior arts. This work, we suppose, paves the way for interactive diffusion-based applications and establishes a foundation for rapid data generation. Code is available at https://github.com/UnicomAI/ShortDF.

Key facts. Cited by 6 papers in the map (data through 13 Oct 2026). Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models. Builds on 30 of its 37 references that are in the map. First posted on arXiv 5 Mar 2025.

Builds on

30 of 37 references.

Closest ancestors · 20

Same direction, newest first

  1. 2023Fast ODE-based Sampling for Diffusion Models in Around 5 Steps30 Nov
  2. 2023One-step Diffusion with Distribution Matching Distillation30 Nov
  3. 2023DPM-Solver-v3: Improved Diffusion ODE Solver with Empirical Model Statistics20 Oct
  4. 2023Distilling ODE Solvers of Diffusion Models into Smaller Steps28 Sep
  5. 2023Spiking Denoising Diffusion Probabilistic Models29 Jun
  6. 2023BOOT: Data-free Distillation of Denoising Diffusion Models with Bootstrapping8 Jun
  7. 2023TRACT: Denoising Diffusion Models with Transitive Closure Time-Distillation7 Mar
  8. 2023Consistency Models2 Mar
  9. 2022Fast Sampling of Diffusion Models via Operator Learning24 Nov
  10. 2022On Distillation of Guided Diffusion Models6 Oct
  11. 2022Diffusion Models: A Comprehensive Survey of Methods and Applications2 Sep
  12. 2022gDDIM: Generalized denoising diffusion implicit models11 Jun
8 more
  1. 2022DPM-Solver: A Fast ODE Solver for Diffusion Probabilistic Model Sampling in Around 10 Steps2 Jun
  2. 2022Elucidating the Design Space of Diffusion-Based Generative Models1 Jun
  3. 2022Fast Sampling of Diffusion Models with Exponential Integrator29 Apr
  4. 2022Pseudo Numerical Methods for Diffusion Models on Manifolds20 Feb
  5. 2021Structured Denoising Diffusion Models in Discrete State-Spaces7 Jul
  6. 2020Denoising Diffusion Implicit Models6 Oct
  7. 2020Denoising Diffusion Probabilistic Models19 Jun
  8. 2020Improved Techniques for Training Score-Based Generative Models16 Jun

Other references · 10

  1. 2014Deep Learning Face Attributes in the Wildcited by 3,686
  2. 2022Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flowcited by 3,607
  3. 2015LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loopcited by 1,350
  4. 2023ProlificDreamer: High-Fidelity and Diverse Text-to-3D Generation with Variational Score Distillationcited by 1,212
  5. 2022Diffusion-LM Improves Controllable Text Generationcited by 1,012
  6. 2021Label-Efficient Semantic Segmentation with Diffusion Modelscited by 496
  7. 2023InstaFlow: One Step is Enough for High-Quality Diffusion-Based Text-to-Image Generationcited by 391
  8. 2022Lossy Image Compression with Conditional Diffusion Modelscited by 168
  9. 2023Residual Denoising Diffusion Modelscited by 68
  10. 2023Resfusion: Denoising Diffusion Probabilistic Models for Image Restoration Based on Prior Residual Noisecited by 19

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2025: 1; 2026: 5.

6 papers.

  1. 2025LeMiCa: Lexicographic Minimax Path Caching for Efficient Diffusion-Based Video Generationcited by 10
  2. 2026MeanCache: From Instantaneous to Average Velocity for Accelerating Flow Matching Inferencecited by 9
  3. 2026Chain-of-Trajectories: Unlocking the Intrinsic Generative Optimality of Diffusion Models via Graph-Theoretic Planningcited by 1
  4. 2026OTCache: Optimal Transport for Geometry-Aware Caching in Diffusion Modelscited by 1
  5. 2026SSPFormer: Self-Supervised Pretrained Transformer for MRI Imagescited by 0
  6. 2026Beyond Geometry: Artistic Disparity Synthesis for Immersive 2D-to-3Dcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2023Image generation with shortest path diffusion
  2. 2023Interpreting and Improving Diffusion Models from an Optimization Perspective
  3. 2026Efficient Coarse-to-Fine Diffusion Models with Time Step Sequence Redistribution
  4. 2023ShiftDDPMs: Exploring Conditional Diffusion Models by Shifting Diffusion Trajectories
  5. 2025You Only Look One Step: Accelerating Backpropagation in Diffusion Sampling with Gradient Shortcuts
  6. 2023Iterative α-(de)Blending: a Minimalist Deterministic Diffusion Model
  7. 2025ShortFT: Diffusion Model Alignment via Shortcut-based Fine-Tuning
  8. 2023The Missing U for Efficient Diffusion Models