Call for Papers
Accepted papers will be presented as posters during the poster sessions. Selected works will also be highlighted as contributed talks.
Openreview venue
Our openreview website for submission is at https://openreview.net/group?id=ICML.cc/2026/Workshop/SPIGM
Topics
We invite submissions related (but not limited) to the following topics:
- Inference and generating methods for graphs, time series, text, video, and other structured modalities
- Unsupervised representation learning of high dimensional structured data
- Sampling and variational inference
- Intersection between probabilistic inference and LLMs, VLMs, VLAs, and foundation models
- Scaling and accelerating inference and generative models on structured data
- Uncertainty quantification in AI systems
- Applications and practical implementations of existing methods to areas in science
- Empirical analysis comparing different architectures for a given data modality and application
Important Dates (Anywhere on Earth)
- Submission:
April 24 AOE, 20261 May AOE, 20268 May AOE, 2026 - Review deadline:
May 10 AOE, 202618 May AOE, 2026 - Accept / Reject notification date:
May 15 AOE, 202622 May AOE, 2026 - Workshop date: July 10 or 11, 2026 (TBD)
We are grateful to ICML for allowing us to extend the notification deadline by one week. However, applicants who wish to apply for ICML financial aid should still submit their financial aid application by 18 May, even if the acceptance status of their workshop paper is not yet known.
Details
- Formatting Instructions: We solicit 4-to-8-page workshop papers (with unlimited references and appendix) following ICML 2026 main conference paper template. The maximum size of submissions is 50 MB. While your submission can contain a supplement or appendix, please note that reviewers are not obliged to review supplementary material.
- Reviews: The review process will be double-blind. All submissions must be anonymized and the leakage of any identification information is prohibited.
- What can be sumitted: Following the convention of most workshops, we welcome submissions of papers already published in journals or presented at non–machine-learning conferences or workshops. If a paper has appeared in a machine-learning venue—such as ICLR, NeurIPS, ICML, CVPR, AISTATS, or similar—we will consider it only if it includes substantial extensions or new results. We also recognize that, due to unforeseen circumstances (e.g., visa), some authors have published papers but did not have the opportunity to present them; accordingly, this year we make an exception to allow those papers to submit to our workshop as well.