# Optimal Inference with Black-box Predictions

arXiv: https://arxiv.org/abs/2608.10155 · HTML: https://mlsanity.com/paper/2608.10155/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 46 references that are in the map.
- First posted on arXiv 10 Aug 2026.

## Builds on (most cited first, 9 of 9)

- [PPI++: Efficient Prediction-Powered Inference](https://mlsanity.com/paper/2311.01453/) (2023)
- [Semi-Supervised Quantile Estimation: Robust and Efficient Inference in High Dimensional Settings](https://mlsanity.com/paper/2201.10208/) (2022)
- [A General Framework for Treatment Effect Estimation in Semi-Supervised and High Dimensional Settings](https://mlsanity.com/paper/2201.00468/) (2022)
- [Doubly-robust inference and optimality in structure-agnostic models with smoothness](https://mlsanity.com/paper/2405.08525/) (2024)
- [Structure-agnostic Optimality of Doubly Robust Learning for Treatment Effect Estimation](https://mlsanity.com/paper/2402.14264/) (2024)
- [Sharp Structure-Agnostic Lower Bounds for General Linear Functional Estimation](https://mlsanity.com/paper/2512.17341/) (2025)
- [Generalized Prediction-Powered Inference, with Application to Binary Classifier Evaluation](https://mlsanity.com/paper/2602.10332/) (2026)
- [Honest Inference for Stochastic Optimization](https://mlsanity.com/paper/2501.07772/) (2025)
- [Do More Predictions Improve Statistical Inference? Filtered Prediction-Powered Inference](https://mlsanity.com/paper/2602.10464/) (2026)

## Abstract (arXiv)

Powerful black-box predictive models have motivated many proposals for combining observed data with predictions to perform valid statistical inference. Despite this progress, the field lacks a unifying principle that explains how hypothesis tests should integrate data and predictions in a way that is both valid and efficient. In this work, we address this gap in the high-dimensional Gaussian sequence model. We characterize the information-theoretic limits of inference with black-box predictions when their accuracies are known and, for orthogonal predictions, when they are unknown. Building on these characterizations, we develop practical hypothesis tests that adapt to the unknown accuracies of the predictions while benefiting from strong alignment among them.