REVIEW 10 cited by
Prediction-Powered Inference
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Prediction-Powered Inference
read the original abstract
Prediction-powered inference is a framework for performing valid statistical inference when an experimental dataset is supplemented with predictions from a machine-learning system. The framework yields simple algorithms for computing provably valid confidence intervals for quantities such as means, quantiles, and linear and logistic regression coefficients, without making any assumptions on the machine-learning algorithm that supplies the predictions. Furthermore, more accurate predictions translate to smaller confidence intervals. Prediction-powered inference could enable researchers to draw valid and more data-efficient conclusions using machine learning. The benefits of prediction-powered inference are demonstrated with datasets from proteomics, astronomy, genomics, remote sensing, census analysis, and ecology.
Forward citations
Cited by 10 Pith papers
-
The Statistical Cost of Adaptation in Multi-Source Transfer Learning
Multi-source transfer learning incurs an intrinsic adaptation cost that can exceed one, with phase transitions separating regimes where bias-agnostic estimators match oracle performance from those where they cannot.
-
NaiAD: Initiate Data-Driven Research for LLM Advertising
NaiAD is a new dataset and framework for LLM-native advertising that uses decoupled generation and calibrated scoring to identify four semantic strategies for balancing user and commercial utilities.
-
Reasoning Gets Harder for LLMs Inside A Dialogue
LLMs show a consistent performance drop on arithmetic, spatial, and temporal reasoning tasks when framed in multi-turn dialogues versus isolated settings, demonstrated by the new BOULDER benchmark across eight travel-...
-
Power-Optimal Covariate Adjustment for Switchback Experiments
Training the CUPAC covariate and its regression coefficient with a between-cell-weighted loss improves switchback estimator power, with gains concentrated in within-noise-dominated regimes.
-
Real vs. Semi-Simulated: Rethinking Evaluation for Treatment Effect Estimation
Counterfactual metrics on semi-simulated benchmarks fail to identify the treatment effect estimators preferred by observable metrics on real datasets, with simple meta-learners outperforming specialized causal models.
-
Real vs. Semi-Simulated: Rethinking Evaluation for Treatment Effect Estimation
Large-scale study finds that counterfactual metrics on semi-simulated data do not select the same estimators as observable metrics on real data, and benchmark rankings fail to transfer.
-
Response Time Enhances Alignment with Heterogeneous Preferences
Response times modeled as drift-diffusion processes enable consistent estimation of population-average preferences from heterogeneous anonymous binary choices.
-
PPI++: Efficient Prediction-Powered Inference
PPI++ yields easy-to-compute confidence sets for any-dimensional parameters that always improve on classical intervals from labeled data alone by leveraging abundant ML predictions.
-
Efficient Sequential Evaluation of Large Language Models
A confidence-sequence framework for sequentially estimating an LLM's average benchmark accuracy under adaptive question selection, with growth-oriented sampling rules that in practice often lose to uniform sampling.
-
Towards Auditing AI Systems in the Wild
Proposes framing auditing of deployed AI systems as continuous statistical monitoring of risk-controlled constraints like fairness and safety under uncertainty.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.