mlsanityData to 7 Oct 2026

PPI++: Efficient Prediction-Powered Inference

arXiv 2311.01453 ↗First posted 2 Nov 2023stat.ML cs.LG stat.MEcited by 103

Not on the map. Why?

Abstract (arXiv)

We present PPI++: a computationally lightweight methodology for estimation and inference based on a small labeled dataset and a typically much larger dataset of machine-learning predictions. The methods automatically adapt to the quality of available predictions, yielding easy-to-compute confidence sets -- for parameters of any dimensionality -- that always improve on classical intervals using only the labeled data. PPI++ builds on prediction-powered inference (PPI), which targets the same problem setting, improving its computational and statistical efficiency. Real and synthetic experiments demonstrate the benefits of the proposed adaptations.

Key facts. Cited by 103 papers in the map (data through 7 Oct 2026). Builds on 9 of its 33 references that are in the map. First posted on arXiv 2 Nov 2023.

Builds on

9 of 33 references.

References · 9

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2016XGBoost: A Scalable Tree Boosting Systemcited by 5,120
  3. 2021Convex optimizationcited by 2,479
  4. 2021Retiring Adult: New Datasets for Fair Machine Learningcited by 443
  5. 2023Prediction-Powered Inferencecited by 214
  6. 2019Veridical Data Sciencecited by 65
  7. 2023Cross-Prediction-Powered Inferencecited by 57
  8. 2016Semi-supervised Inference: General Theory and Estimation of Meanscited by 38
  9. 2019High-dimensional semi-supervised learning: in search for optimal inference of the meancited by 8

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 2; 2024: 16; 2025: 37; 2026: 48.

103 papers.

Sort:

By area:

  1. 2024AutoEval Done Right: Using Synthetic Data for Model Evaluationcited by 53
  2. 2024Active Statistical Inferencecited by 46
  3. 2024Limits to scalable evaluation at the frontier: LLM as Judge won't beat twice the datacited by 46
  4. 2025Predictions as Surrogates: Revisiting Surrogate Outcomes in the Age of AIcited by 31
  5. 2024Prediction-Powered Ranking of Large Language Modelscited by 29
  6. 2025How Benchmark Prediction from Fewer Data Misses the Markcited by 26
  7. 2023Assumption-Lean and Data-Adaptive Post-Prediction Inferencecited by 25
  8. 2025How to Correctly Report LLM-as-a-Judge Evaluationscited by 25
  9. 2024Stratified Prediction-Powered Inference for Hybrid Language Model Evaluationcited by 24
  10. 2026Noisy but Valid: Robust Statistical Evaluation of LLMs with Imperfect Judgescited by 23
  11. 2025Prediction-Powered E-Valuescited by 16
  12. 2025Cost-Optimal Active AI Model Evaluationcited by 14
38 more
  1. 2023Prediction De-Correlated Inference: A safe approach for post-prediction inferencecited by 13
  2. 2024A Framework for Efficient Model Evaluation through Stratification, Sampling, and Estimationcited by 13
  3. 2024Semi-Supervised Risk Control via Prediction-Powered Inferencecited by 13
  4. 2025Multi-Armed Bandits With Machine Learning-Generated Surrogate Rewardscited by 13
  5. 2025Reliable and Scalable Robot Policy Evaluation with Imperfect Simulatorscited by 13
  6. 2024A Note on the Prediction-Powered Bootstrapcited by 10
  7. 2025Efficient Randomized Experiments Using Foundation Modelscited by 10
  8. 2025Prediction-Powered Causal Inferencescited by 9
  9. 2025Adaptive Prediction-Powered AutoEval with Reliability and Efficiency Guaranteescited by 9
  10. 2026Multiple-Prediction-Powered Inferencecited by 9
  11. 2024Task-Agnostic Machine-Learning-Assisted Inferencecited by 8
  12. 2025No Free Lunch: Non-Asymptotic Analysis of Prediction-Powered Inferencecited by 8
  13. 2025FAB-PPI: Frequentist, Assisted by Bayes, Prediction-Powered Inferencecited by 7
  14. 2025Anytime-valid, Bayes-assisted, Prediction-Powered Inferencecited by 7
  15. 2025Probably Approximately Correct Labelscited by 7
  16. 2025Valid Inference with Imperfect Synthetic Datacited by 7
  17. 2025Doubly-Robust LLM-as-a-Judge: Externally Valid Estimation with Imperfect Personascited by 7
  18. 2026Efficient Inference for Noisy LLM-as-a-Judge Evaluationcited by 7
  19. 2024Local Prediction-Powered Inferencecited by 6
  20. 2024Regression for the Mean: Auto-Evaluation and Inference with Few Labels through Post-hoc Regressioncited by 6
  21. 2024Another look at statistical inference with machine learning-imputed datacited by 6
  22. 2024Constructing Confidence Intervals for Average Treatment Effects from Multiple Datasetscited by 6
  23. 2025Accelerating Unbiased LLM Evaluation via Synthetic Feedbackcited by 6
  24. 2025AI-Powered Bayesian Inferencecited by 6
  25. 2026PPI is the Difference Estimator: Recognizing the Survey Sampling Roots of Prediction-Powered Inferencecited by 6
  26. 2025Generalization Bounds and Stopping Rules for Learning with Self-Selected Datacited by 5
  27. 2025A Unified Framework for Inference with General Missingness Patterns and Machine Learning Imputationcited by 5
  28. 2025General Synthetic-Powered Inferencecited by 5
  29. 2025Prediction-Powered Semi-Supervised Learning with Online Power Tuningcited by 5
  30. 2025Robust Sampling for Active Statistical Inferencecited by 5
  31. 2025Statistical Inference under Performativitycited by 4
  32. 2025How can we assess human-agent interactions? Case studies in software agent designcited by 4
  33. 2026Demystifying Prediction Powered Inferencecited by 4
  34. 2026A Statistical Framework for Alignment with Biased AI Feedbackcited by 4
  35. 2024Automatic AI Model Selection for Wireless Systems: Online Learning via Digital Twinningcited by 3
  36. 2025ACT as Human: Multimodal Large Language Model Data Annotation with Critical Thinkingcited by 3
  37. 2026PPI-SVRG: Unifying Prediction-Powered Inference and Variance Reduction for Semi-Supervised Optimizationcited by 3
  38. 2026Revisiting Active Sequential Prediction-Powered Mean Estimationcited by 3

50 most cited of 103 All 103 (JSON)