We invite all submissions of original work across three different tracks:

1. Research Papers

2. Challenges and Reflections

3. Benchmarks and Datasets

These submissions should be a maximum of 4 pages in length (excluding references and appendices). Please format your submission using the NeurIPS 2025 Latex style file, which includes an option for workshops with double-blind reviewing format (dblblindworkshop). Use "Frontiers in Probabilistic Inference: Sampling Meets Learning" as the workshop title.

All submissions are non-archival, and must present original work that has not been previously accepted at any archival venue at the time of submission.

Submissions should be anonymized and will undergo a single round of double-blind peer review.

You can submit your work via OpenReview, selecting the Main Track for your submission.