When Your Model Stops Working: Anytime-Valid Calibration Monitoring

• URL: http://arxiv.org/abs/2603.13156v1
• Date: Fri, 13 Mar 2026 16:50:14 GMT
• Title: When Your Model Stops Working: Anytime-Valid Calibration Monitoring
• Abstract summary: PITMonitor is an anytime-valid calibration-specific monitor that detects distributional shifts in probability integral transforms.<n>On river's FriedmanDrift benchmark, PITMonitor detection rates achieves competitive with the strongest baselines.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Practitioners monitoring deployed probabilistic models face a fundamental trap: any fixed-sample test applied repeatedly over an unbounded stream will eventually raise a false alarm, even when the model remains perfectly stable. Existing methods typically lack formal error guarantees, conflate alarm time with changepoint location, and monitor indirect signals that do not fully characterize calibration. We present PITMonitor, an anytime-valid calibration-specific monitor that detects distributional shifts in probability integral transforms via a mixture e-process, providing Type I error control over an unbounded monitoring horizon as well as Bayesian changepoint estimation. On river's FriedmanDrift benchmark, PITMonitor achieves detection rates competitive with the strongest baselines across all three scenarios, although detection delay is substantially longer under local drift.

Related papers

• Beyond Uncertainty: Generalizable Failure Monitoring for Surgical Segmentation under Acquisition Degradation [2.287641923890767]
  Surgical segmentation networks can fail silently under acquisition degradation.<n>We present TCSR-Monitor, a post-hoc failure-monitoring framework that combines confidence with observable shape, temporal-consistency, and image-quality cues.
  arXiv  Detail & Related papers  (2026-08-17T15:57:19Z)
• Discovery of Hidden Miscalibration Regimes [52.452902154360565]
  Models may be systematically overconfident on some kinds of inputs and underconfident on others.<n>We define the corresponding miscalibration field and propose a diagnostic framework for estimating it.<n>Our approach learns a calibration-aware representation of the input space and estimates signed local miscalibration by kernel smoothing in the learned geometry.
  arXiv  Detail & Related papers  (2026-05-13T13:07:50Z)
• Distributional Spectral Diagnostics for Localizing Grokking Transitions [4.480801209867367]
  In grokking, a model first fits the training data while test accuracy remains low, and only later begins to generalize.<n>We ask whether this transition can be localized from observed training trajectories before the test accuracy rises, and formulate grokking transition localization as a diagnostic problem with an explicit threshold/FPR/lead-time trade-off.
  arXiv  Detail & Related papers  (2026-05-07T03:08:58Z)
• Testing For Distribution Shifts with Conditional Conformal Test Martingales [20.369245689839822]
  We propose a sequential test for detecting arbitrary distribution shifts that allows conformal test martingales (CTMs) to work under a fixed, reference-conditional setting.<n>Existing CTM detectors construct test martingales by continually growing a reference set with each incoming sample, using it to assess how atypical the new sample is relative to past observations.<n>While this design yields anytime-valid type-I error control, it suffers from test-time contamination: after a change, post-shift observations enter the reference set and dilute the evidence for distribution shift, increasing detection delay and reducing power.<n>Our method detects shifts faster than standard
  arXiv  Detail & Related papers  (2026-02-14T18:47:26Z)
• Technical note on Sequential Test-Time Adaptation via Martingale-Driven Fisher Prompting [3.5808917363708743]
  M-FISHER is a method for sequential distribution shift detection and stable adaptation in streaming data.<n>For detection, we construct an exponential martingale from non-conformity scores and apply Ville's inequality to obtain time-uniform guarantees on false alarm control.<n>For adaptation, we show that Fisher-preconditioned updates of prompt parameters implement natural gradient descent on the distributional manifold.
  arXiv  Detail & Related papers  (2025-10-04T15:31:26Z)
• Revisiting Multivariate Time Series Forecasting with Missing Values [65.30332997607141]
  Missing values are common in real-world time series.<n>Current approaches have developed an imputation-then-prediction framework that uses imputation modules to fill in missing values, followed by forecasting on the imputed data.<n>This framework overlooks a critical issue: there is no ground truth for the missing values, making the imputation process susceptible to errors that can degrade prediction accuracy.<n>We introduce Consistency-Regularized Information Bottleneck (CRIB), a novel framework built on the Information Bottleneck principle.
  arXiv  Detail & Related papers  (2025-09-27T20:57:48Z)
• WATCH: Adaptive Monitoring for AI Deployments via Weighted-Conformal Martingales [22.789611187514975]
  Methods for nonparametric sequential testing -- especially conformal test martingales (CTMs) and anytime-valid inference -- offer promising tools for this monitoring task.<n>Existing approaches are restricted to monitoring limited hypothesis classes or alarm criteria''
  arXiv  Detail & Related papers  (2025-05-07T17:53:47Z)
• Deployment Prior Injection for Run-time Calibratable Object Detection [58.636806402337776]
  We introduce an additional graph input to the detector, where the graph represents the deployment context prior. During the test phase, any suitable deployment context prior can be injected into the detector via graph edits. Even if the deployment prior is unknown, the detector can self-calibrate using deployment prior approximated using its own predictions.
  arXiv  Detail & Related papers  (2024-02-27T04:56:04Z)
• Tracking the risk of a deployed model and detecting harmful distribution shifts [105.27463615756733]
  In practice, it may make sense to ignore benign shifts, under which the performance of a deployed model does not degrade substantially. We argue that a sensible method for firing off a warning has to both (a) detect harmful shifts while ignoring benign ones, and (b) allow continuous monitoring of model performance without increasing the false alarm rate.
  arXiv  Detail & Related papers  (2021-10-12T17:21:41Z)
• Training on Test Data with Bayesian Adaptation for Covariate Shift [96.3250517412545]
  Deep neural networks often make inaccurate predictions with unreliable uncertainty estimates. We derive a Bayesian model that provides for a well-defined relationship between unlabeled inputs under distributional shift and model parameters. We show that our method improves both accuracy and uncertainty estimation.
  arXiv  Detail & Related papers  (2021-09-27T01:09:08Z)
• Detecting Rewards Deterioration in Episodic Reinforcement Learning [63.49923393311052]
  In many RL applications, once training ends, it is vital to detect any deterioration in the agent performance as soon as possible. We consider an episodic framework, where the rewards within each episode are not independent, nor identically-distributed, nor Markov. We define the mean-shift in a way corresponding to deterioration of a temporal signal (such as the rewards), and derive a test for this problem with optimal statistical power.
  arXiv  Detail & Related papers  (2020-10-22T12:45:55Z)

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.