Accepted papers will be presented as posters during the workshop poster session on 2nd December, 2025.
Please be aware that this is not the same as the ELLIS UnConference poster session.
Important Dates (Anywhere on Earth)
Submission due: Oct 16, 2025
Accept/reject notice: Oct 31, 2025
Topics
We invite submissions related (but not limited) to the following topics:
Amortized approaches to simulation-based inference (SBI), e.g., neural posterior, likelihood, or ratio estimators
Amortized / efficient probabilistic methods
Amortized inference / meta-learning for efficient probabilistic methods
Neural processes / PFNs
Amortized sequential decision making (e.g., experimental design, active learning, Bayesian optimization)
Foundation models for inference and design
Applications enabled by amortized probabilistic methods
We are particularly interested in submissions that tackle open questions for amortized ProbML, which include (but are not limited to):
Robustness in amortized inference or amortized sequential decision making (e.g., sim2real, distribution shift, model misspecification)
Post-computation diagnostics and uncertainty assessment (calibration/coverage)
Data/compute/latency trade-offs and scaling laws
Going beyond pure pretraining (e.g., test-time compute)
Submission Instruction
Submissions should be made via OpenReview
Please submit approximately half-page (plus unlimited space for references), anonymized abstracts.
Note OpenReview policy:
New profiles created with an institutional email will be activated automatically.
New profiles created without an institutional email will go through a moderation process that can take up to two weeks.