mlsanityData to 14 Oct 2026

Synthetic-Powered Predictive Inference

arXiv 2505.13432 ↗First posted 19 May 2025cs.LG stat.MLcited by 9

Not on the map. Why?

Abstract (arXiv)

Conformal prediction is a framework for predictive inference with a distribution-free, finite-sample guarantee. However, it tends to provide uninformative prediction sets when calibration data are scarce. This paper introduces Synthetic-powered predictive inference (SPI), a novel framework that incorporates synthetic data -- e.g., from a generative model -- to improve sample efficiency. At the core of our method is a score transporter: an empirical quantile mapping that aligns nonconformity scores from trusted, real data with those from synthetic data. By carefully integrating the score transporter into the calibration process, SPI provably achieves finite-sample coverage guarantees without making any assumptions about the real and synthetic data distributions. When the score distributions are well aligned, SPI yields substantially tighter and more informative prediction sets than standard conformal prediction. Experiments on image classification -- augmenting data with synthetic diffusion-model generated images -- and on tabular regression demonstrate notable improvements in predictive efficiency in data-scarce settings.

Key facts. Cited by 9 papers in the map (data through 14 Oct 2026). Builds on 39 of its 58 references that are in the map. First posted on arXiv 19 May 2025.

Builds on

39 of 58 references.

References · 39

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  3. 2022LAION-5B: An open large-scale dataset for training next generation image-text modelscited by 4,152
  4. 2021On the Opportunities and Risks of Foundation Modelscited by 4,033
  5. 2021A Gentle Introduction to Conformal Prediction and Distribution-Free Uncertainty Quantificationcited by 792
  6. 2016Distribution-Free Predictive Inference For Regressioncited by 668
  7. 2019Conformalized Quantile Regressioncited by 629
  8. 2019Conformal Prediction Under Covariate Shiftcited by 519
  9. 2020Classification with Valid and Adaptive Coveragecited by 396
  10. 2021Adaptive Conformal Inference Under Distribution Shiftcited by 375
  11. 2022Conformal prediction beyond exchangeabilitycited by 355
  12. 2016Least Ambiguous Set-Valued Classifiers with Bounded Error Levelscited by 272
27 more
  1. 2023Prediction-Powered Inferencecited by 220
  2. 2023Conformal Prediction With Conditional Guaranteescited by 152
  3. 2021Testing for Outliers with Conformal p-valuescited by 134
  4. 2021Localized Conformal Prediction: A Generalized Inference Framework for Conformal Predictioncited by 114
  5. 2023Class-Conditional Conformal Prediction with Many Classescited by 109
  6. 2021Distribution-free uncertainty quantification for classification under label shiftcited by 101
  7. 2013A Conformal Prediction Approach to Explore Functional Datacited by 86
  8. 2022Batch Multivalid Conformal Predictioncited by 67
  9. 2018Exact and Robust Conformal Inference Methods for Predictive Machine Learning With Dependent Datacited by 66
  10. 2022Training Uncertainty-Aware Classifiers with Conformalized Deep Learningcited by 64
  11. 2021Few-shot Conformal Prediction with Auxiliary Taskscited by 53
  12. 2023Conformal prediction with local weights: randomization enables local guaranteescited by 48
  13. 2021PAC Prediction Sets Under Covariate Shiftcited by 45
  14. 2019Prediction and outlier detection in classification problemscited by 44
  15. 2018Distribution-Free Prediction Sets for Two-Layer Hierarchical Modelscited by 35
  16. 2023Conformal prediction under ambiguous ground truthcited by 31
  17. 2024Boosted Conformal Prediction Intervalscited by 27
  18. 2023Adaptive conformal classification with noisy labelscited by 25
  19. 2022Integrative conformal p-values for powerful out-of-distribution testing with labeled outlierscited by 19
  20. 2022Few-Shot Calibration of Set Predictors via Meta-Learned Cross-Validation-Based Conformal Predictioncited by 17
  21. 2023Conformal inference is (almost) free for neural networks trained with early stoppingcited by 14
  22. 2024Semi-Supervised Risk Control via Prediction-Powered Inferencecited by 13
  23. 2025Kandinsky Conformal Prediction: Beyond Class- and Covariate-Conditional Coveragecited by 13
  24. 2023Estimating Uncertainty in Multimodal Foundation Models using Public Internet Datacited by 10
  25. 2024Batch Predictive Inferencecited by 7
  26. 2024Posterior Conformal Predictioncited by 7
  27. 2023A conformal test of linear models via permutation-augmented regressionscited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 3; 2026: 6.

9 papers.

Similar, not cited

Close in wording, no citation either way

  1. 2026Sample-Efficient Generative Conformal Prediction
  2. 2025Boosting Statistic Learning with Synthetic Data from Pretrained Large Models
  3. 2024Provably Reliable Conformal Prediction Sets in the Presence of Data Poisoning
  4. 2023How to Trust Your Diffusion Model: A Convex Optimization Approach to Conformal Risk Control
  5. 2025Unsupervised Conformal Inference: Bootstrapping and Alignment to Control LLM Uncertainty
  6. 2024Optimized Conformal Selection: Powerful Selective Inference After Conformity Score Optimization
  7. 2025CP4SBI: Local Conformal Calibration of Credible Sets in Simulation-Based Inference
  8. 2026Empirical Bayes Conformal Prediction for Vision and Language Models