Skip to main content
archive
Search Submit Donate Log in
Press Enter to search · Advanced search

Computer Science > Computer Vision and Pattern Recognition

arXiv:2305.15583 (cs)
[Submitted on 24 May 2023 (v1), last revised 16 Jun 2024 (this version, v8)]

Title:Alleviating Exposure Bias in Diffusion Models through Sampling with Shifted Time Steps

Authors:Mingxiao Li, Tingyu Qu, Ruicong Yao, Wei Sun, Marie-Francine Moens
View a PDF of the paper titled Alleviating Exposure Bias in Diffusion Models through Sampling with Shifted Time Steps, by Mingxiao Li and 4 other authors
View PDF HTML (experimental)
Abstract:Diffusion Probabilistic Models (DPM) have shown remarkable efficacy in the synthesis of high-quality images. However, their inference process characteristically requires numerous, potentially hundreds, of iterative steps, which could exaggerate the problem of exposure bias due to the training and inference discrepancy. Previous work has attempted to mitigate this issue by perturbing inputs during training, which consequently mandates the retraining of the DPM. In this work, we conduct a systematic study of exposure bias in DPM and, intriguingly, we find that the exposure bias could be alleviated with a novel sampling method that we propose, without retraining the model. We empirically and theoretically show that, during inference, for each backward time step $t$ and corresponding state $\hat{x}_t$, there might exist another time step $t_s$ which exhibits superior coupling with $\hat{x}_t$. Based on this finding, we introduce a sampling method named Time-Shift Sampler. Our framework can be seamlessly integrated to existing sampling algorithms, such as DDPM, DDIM and other high-order solvers, inducing merely minimal additional computations. Experimental results show our method brings significant and consistent improvements in FID scores on different datasets and sampling methods. For example, integrating Time-Shift Sampler to F-PNDM yields a FID=3.88, achieving 44.49\% improvements as compared to F-PNDM, on CIFAR-10 with 10 sampling steps, which is more performant than the vanilla DDIM with 100 sampling steps. Our code is available at this https URL.
Comments: Accepted at International Conference on Learning Representations (ICLR2024); typo correction
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Cite as: arXiv:2305.15583 [cs.CV]
  (or arXiv:2305.15583v8 [cs.CV] for this version)
  https://doi.org/10.48550/arXiv.2305.15583
arXiv-issued DOI via DataCite

Submission history

From: Tingyu Qu [view email]
[v1] Wed, 24 May 2023 21:39:27 UTC (8,568 KB)
[v2] Fri, 26 May 2023 08:17:35 UTC (8,568 KB)
[v3] Sat, 30 Sep 2023 09:19:37 UTC (9,300 KB)
[v4] Tue, 5 Dec 2023 13:48:12 UTC (10,710 KB)
[v5] Wed, 17 Jan 2024 12:13:50 UTC (10,710 KB)
[v6] Thu, 14 Mar 2024 14:15:48 UTC (10,710 KB)
[v7] Thu, 21 Mar 2024 14:16:39 UTC (10,710 KB)
[v8] Sun, 16 Jun 2024 15:12:11 UTC (10,710 KB)
Full-text links:

Access Paper:

    View a PDF of the paper titled Alleviating Exposure Bias in Diffusion Models through Sampling with Shifted Time Steps, by Mingxiao Li and 4 other authors
  • View PDF
  • HTML (experimental)
  • TeX Source
view license

Current browse context:

cs.CV
< prev   |   next >
new | recent | 2023-05
Change to browse by:
cs

References & Citations

  • NASA ADS
  • Google Scholar
  • Semantic Scholar
Loading...

BibTeX formatted citation

Data provided by:

Bookmark

BibSonomy Reddit

Bibliographic and Citation Tools

Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)

Code, Data and Media Associated with this Article

alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
ScienceCast (What is ScienceCast?)

Demos

Replicate (What is Replicate?)
Hugging Face Spaces (What is Spaces?)
TXYZ.AI (What is TXYZ.AI?)

Recommenders and Search Tools

Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)
  • Author
  • Venue
  • Institution
  • Topic

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)
We gratefully acknowledge support from our major funders, member institutions, , and all contributors.
About · Help · Contact · Subscribe · Copyright · Privacy · Accessibility · Operational Status (opens in new tab)
Major funding support from
Simons Foundation Simons Foundation International Schmidt Sciences