Prediction-Powered Inference Across Many Tasks for AI Evaluation & Social Science Research

• URL: http://arxiv.org/abs/2605.29249v1
• Date: Thu, 28 May 2026 02:09:38 GMT
• Title: Prediction-Powered Inference Across Many Tasks for AI Evaluation & Social Science Research
• Abstract summary: Many applications require statistically valid inference across many related tasks, while using only a handful of high-quality labels per hypothesis.<n>We introduce a prediction-powered inference framework that uses labeled data from related tasks to improve power while preserving task-specific inference.<n>We show that cross-task recalibration can substantially reduce confidence interval widths when labels are scarce.
• Score: 6.716363754264257
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Many applications require statistically valid inference across many related tasks, while using only a handful of high-quality labels per hypothesis. In AI evaluation, these tasks may correspond to model behaviors across prompts, subgroups, or hypotheses; in social science surveys, they may correspond to related questions, populations, or measurement conditions. Prediction-powered inference (PPI) uses abundant but inexpensive proxy measurements to improve inference from limited, ground-truth labels, but commonly used methods treat tasks independently and therefore fail to exploit shared structure across related tasks. This limitation is especially important in settings where only a small number of labels are available per task. To address this issue, we introduce a multi-task prediction-powered inference framework that uses labeled data from related tasks to improve power while preserving task-specific inference. Our methods exploit the shared structure in the proxy-ground-truth relationship through cross-task recalibration, while retaining within-task rectification and power tuning to construct accurate point estimates and confidence intervals. We prove that efficiency gains beyond power-tuned PPI are only possible when the proxy-ground-truth relationship contains nonlinear structure; affine cross-task recalibrations are asymptotically equivalent to using the original proxy. We complement our theoretical findings with experiments on synthetic and semi-synthetic datasets, as well as a case study auditing language models on election-related information during the 2024 U.S. presidential election. Using a large human-annotation study, we show that cross-task recalibration can substantially reduce confidence interval widths when labels are scarce.

Related papers

• Causally Evaluating the Learnability of Formal Language Tasks [59.73001113068827]
  We rigorously investigate the relationship between data frequency and learnability.<n>We derive Kullback-Leibler divergence metrics to measure the learnability of specific sub-tasks.<n>Our experiments show that evaluating learnability without causal intervention leads to incorrect conclusions.
  arXiv  Detail & Related papers  (2026-06-08T17:58:36Z)
• On the Paradoxical Interference between Instruction-Following and Task Solving [50.75960598434753]
  Instruction following aims to align Large Language Models (LLMs) with human intent by specifying explicit constraints on how tasks should be performed.<n>We reveal a counterintuitive phenomenon: instruction following can paradoxically interfere with LLMs' task-solving capability.<n>We propose a metric, SUSTAINSCORE, to quantify the interference of instruction following with task solving.
  arXiv  Detail & Related papers  (2026-01-29T17:48:56Z)
• Demystifying Prediction Powered Inference [4.962232906170314]
  Prediction-Powered Inference (PPI) offers a principled framework that leverages predictions from large unlabeled datasets to improve statistical efficiency.<n>Despite its potential, the growing PPI variants and the subtle distinctions between them have made it challenging for practitioners to determine when and how to apply these methods responsibly.<n>This paper demystifies PPI by synthesizing its theoretical foundations, methodological extensions, connections to existing statistics literature, and diagnostic tools into a unified practical workflow.
  arXiv  Detail & Related papers  (2026-01-28T18:16:02Z)
• Generalisation in Multitask Fitted Q-Iteration and Offline Q-learning [0.0]
  We study offline multitask reinforcement learning in settings where multiple tasks share a low-rank representation of their action-value functions.<n>We analyze a multitask variant of fitted Q-iteration that jointly learns a shared representation and task-specific value functions.<n>Our results clarify the role of shared representations in multitask offline Q-learning and provide theoretical insight into when and how multitask structure can improve generalization.
  arXiv  Detail & Related papers  (2025-12-23T10:20:11Z)
• Probably Approximately Precision and Recall Learning [60.00180898830079]
  A key challenge in machine learning is the prevalence of one-sided feedback.<n>We introduce a Probably Approximately Correct (PAC) framework in which hypotheses are set functions that map each input to a set of labels.<n>We develop new algorithms that learn from positive data alone, achieving optimal sample complexity in the realizable case.
  arXiv  Detail & Related papers  (2024-11-20T04:21:07Z)
• IT$^3$: Idempotent Test-Time Training [95.78053599609044]
  Deep learning models often struggle when deployed in real-world settings due to distribution shifts between training and test data.<n>We present Idempotent Test-Time Training (IT$3$), a novel approach that enables on-the-fly adaptation to distribution shifts using only the current test instance.<n>Our results suggest that idempotence provides a universal principle for test-time adaptation that generalizes across domains and architectures.
  arXiv  Detail & Related papers  (2024-10-05T15:39:51Z)
• Selective Inference for Sparse Multitask Regression with Applications in Neuroimaging [2.611153304251067]
  We propose a framework for selective inference to address a common multi-task problem in neuroimaging. Our framework offers a new conditional procedure for inference, based on a refinement of the selection event that yields a tractable selection-adjusted likelihood. We demonstrate through simulations that multi-task learning with selective inference can more accurately recover true signals than single-task methods.
  arXiv  Detail & Related papers  (2022-05-27T20:21:20Z)
• Learning with Instance Bundles for Reading Comprehension [61.823444215188296]
  We introduce new supervision techniques that compare question-answer scores across multiple related instances. Specifically, we normalize these scores across various neighborhoods of closely contrasting questions and/or answers. We empirically demonstrate the effectiveness of training with instance bundles on two datasets.
  arXiv  Detail & Related papers  (2021-04-18T06:17:54Z)
• Mediators in Determining what Processing BERT Performs First [35.32217580058933]
  We consider the prediction's context length, namely the length of the span whose processing is minimally required to perform the prediction. We show that not controlling for context length may lead to contradictory conclusions as to the localization patterns of the network.
  arXiv  Detail & Related papers  (2021-04-13T17:58:52Z)
• Robust Learning Through Cross-Task Consistency [92.42534246652062]
  We propose a broadly applicable and fully computational method for augmenting learning with Cross-Task Consistency. We observe that learning with cross-task consistency leads to more accurate predictions and better generalization to out-of-distribution inputs.
  arXiv  Detail & Related papers  (2020-06-07T09:24:33Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.