mlsanityData to 14 Oct 2026

Pre-Trained AI Model Assisted Online Decision-Making under Missing Covariates: A Theoretical Perspective

arXiv 2507.07852 ↗First posted 10 Jul 2025cs.LG stat.MLcited by 3

Not on the map. Why?

Abstract (arXiv)

We study a sequential contextual decision-making problem in which certain covariates are missing but can be imputed using a pre-trained AI model. From a theoretical perspective, we analyze how the presence of such a model influences the regret of the decision-making process. We introduce a novel notion called "model elasticity", which quantifies the sensitivity of the reward function to the discrepancy between the true covariate and its imputed counterpart. This concept provides a unified way to characterize the regret incurred due to model imputation, regardless of the underlying missingness mechanism. More surprisingly, we show that under the missing at random (MAR) setting, it is possible to sequentially calibrate the pre-trained model using tools from orthogonal statistical learning and doubly robust regression. This calibration significantly improves the quality of the imputed covariates, leading to much better regret guarantees. Our analysis highlights the practical value of having an accurate pre-trained model in sequential decision-making tasks and suggests that model elasticity may serve as a fundamental metric for understanding and improving the integration of pre-trained models in a wide range of data-driven decision-making problems.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 33 of its 63 references that are in the map. First posted on arXiv 10 Jul 2025.

Builds on

33 of 63 references.

References · 33

  1. 2019CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparisoncited by 1,423
  2. 2020Meta Pseudo Labelscited by 374
  3. 2020Beyond UCB: Optimal and Efficient Contextual Bandits with Regression Oraclescited by 197
  4. 2023Does Synthetic Data Generation of LLMs Help Clinical Text Mining?cited by 154
  5. 2020Bypassing the Monster: A Faster and Simpler Optimal Algorithm for Contextual Bandits under Realizabilitycited by 115
  6. 2021Adapting to Misspecification in Contextual Banditscited by 81
  7. 2025Synthetic Data Generation Using Large Language Models: Advances in Text and Codecited by 61
  8. 2021Misspecified Gaussian Process Bandit Optimizationcited by 55
  9. 2022Data Augmentation for Intent Classification with Off-the-shelf Large Language Modelscited by 49
  10. 2021Efficient and Optimal Algorithms for Contextual Dueling Bandits under Realizabilitycited by 44
  11. 2016Double/Debiased Machine Learning for Treatment and Causal Parameterscited by 43
  12. 2024LOLA: LLM-Assisted Online Learning Algorithm for Content Experimentscited by 18
21 more
  1. 2023Automatic Debiased Machine Learning for Covariate Shiftscited by 16
  2. 2025Collaborative Prediction: Tractable Information Aggregation via Agreementcited by 15
  3. 2020Offline Contextual Multi-armed Bandits for Mobile Health Interventions: A Case Study on Emotion Regulationcited by 13
  4. 2024Tractable Agreement Protocolscited by 13
  5. 2022Optimal Contextual Bandits with Knapsacks under Realizability via Regression Oraclescited by 12
  6. 2024Semi-Supervised Semantic Segmentation Based on Pseudo-Labels: A Surveycited by 12
  7. 2021On Dynamic Pricing with Covariatescited by 10
  8. 2024Active Exploration via Autoregressive Generation of Missing Datacited by 9
  9. 2020Contextual User Browsing Bandits for Large-Scale Online Mobile Recommendationcited by 8
  10. 2024Offline Oracle-Efficient Learning for Contextual MDPs via Layerwise Exploration-Exploitation Tradeoffcited by 8
  11. 2025The Application of Large Language Models in Recommendation Systemscited by 8
  12. 2025Statistically Valid Post-Deployment Monitoring Should Be Standard for AI-Based Digital Healthcited by 8
  13. 2020Contextual Bandit with Missing Rewardscited by 6
  14. 2025Contextual Thompson Sampling via Generation of Missing Datacited by 6
  15. 2025Contextual Online Decision Making with Infinite-Dimensional Functional Regressioncited by 5
  16. 2023LLM aided semi-supervision for Extractive Dialog Summarizationcited by 4
  17. 2025Dynamic Pricing in the Linear Valuation Model using Shape Constraintscited by 4
  18. 2025Linear Bandits with Partially Observable Featurescited by 4
  19. 2022High dimensional stochastic linear contextual bandit with missing covariatescited by 2
  20. 2023Effective Dimension in Bandit Problems under Censorshipcited by 2
  21. 2024HR-Bandit: Human-AI Collaborated Linear Recourse Banditcited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 2; 2026: 1.

3 papers.

  1. 2026LLM-Inspired Pretrain-Then-Finetune for Small-Data, Large-Scale Optimizationcited by 1
  2. 2025Perturbing the Derivative: Wild Refitting for Model-Free Evaluation of Machine Learning Models under Bregman Lossescited by 0
  3. 2025Learning with Incomplete Context: Linear Contextual Bandits with Pretrained Imputationcited by 0