Toward Calibrated Mixture-of-Experts Under Distribution Shift

• URL: http://arxiv.org/abs/2606.20544v1
• Date: Thu, 18 Jun 2026 17:55:00 GMT
• Title: Toward Calibrated Mixture-of-Experts Under Distribution Shift
• Abstract summary: We show that expert calibration is sufficient to ensure calibration of the overall model under a broad class of distribution shifts.<n>We propose an adversarial reweighting that penalizes calibration errors of the routed aggregate under distribution shift.
• Score: 43.193848309099614
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Calibration aligns a model's predictive uncertainty with the frequencies of its empirical outcomes and is important for understanding and trusting reported probabilities. Recent work shows that enforcing calibration at the level of individual predictors can improve ensemble accuracy and calibration, with mixture-of-experts (MoE) models showing strong empirical improvements in particular; however, the conditions under which calibration helps MoE are not well understood. In this work, we study how MoE models behave under distribution shift, focusing on how routing mechanisms interact with expert-level calibration. We show that expert calibration is sufficient to ensure calibration of the overall model under a broad class of distribution shifts in hard-routed models, but is insufficient for calibrating soft-routed models. To address this, we propose an adversarial reweighting that penalizes calibration errors of the routed aggregate under distribution shift, and we demonstrate that it improves the accuracy-calibration tradeoff both on average and on difficult subsets of the data, across model classes, prediction tasks, and distribution shifts.

Related papers

• CalibrateMix: Guided-Mixup Calibration of Image Semi-Supervised Models [49.588973929678765]
  CalibrateMix is a mixup-based approach that aims to improve the calibration of SSL models.<n>Our method achieves lower expected calibration error (ECE) and superior accuracy compared to existing SSL approaches.
  arXiv  Detail & Related papers  (2025-11-17T04:43:53Z)
• Beyond One-Hot Labels: Semantic Mixing for Model Calibration [22.39558434131574]
  We present textbfCalibration-aware Semantic Mixing (CSM), a novel framework that generates training samples with mixed class characteristics.<n>We show that CSM achieves superior calibration compared to the state-of-the-art calibration approaches.
  arXiv  Detail & Related papers  (2025-04-18T08:26:18Z)
• Reassessing How to Compare and Improve the Calibration of Machine Learning Models [7.183341902583164]
  A machine learning model is calibrated if its predicted probability for an outcome matches the observed frequency for that outcome conditional on the model prediction.<n>We show that there exist trivial recalibration approaches that can appear seemingly state-of-the-art unless calibration and prediction metrics are accompanied by additional generalization metrics.
  arXiv  Detail & Related papers  (2024-06-06T13:33:45Z)
• Calibrating Large Language Models with Sample Consistency [76.23956851098598]
  We explore the potential of deriving confidence from the distribution of multiple randomly sampled model generations, via three measures of consistency. Results show that consistency-based calibration methods outperform existing post-hoc approaches. We offer practical guidance on choosing suitable consistency metrics for calibration, tailored to the characteristics of various LMs.
  arXiv  Detail & Related papers  (2024-02-21T16:15:20Z)
• Calibration by Distribution Matching: Trainable Kernel Calibration Metrics [56.629245030893685]
  We introduce kernel-based calibration metrics that unify and generalize popular forms of calibration for both classification and regression. These metrics admit differentiable sample estimates, making it easy to incorporate a calibration objective into empirical risk minimization. We provide intuitive mechanisms to tailor calibration metrics to a decision task, and enforce accurate loss estimation and no regret decisions.
  arXiv  Detail & Related papers  (2023-10-31T06:19:40Z)
• On Calibrating Semantic Segmentation Models: Analyses and An Algorithm [51.85289816613351]
  We study the problem of semantic segmentation calibration. Model capacity, crop size, multi-scale testing, and prediction correctness have impact on calibration. We propose a simple, unifying, and effective approach, namely selective scaling.
  arXiv  Detail & Related papers  (2022-12-22T22:05:16Z)
• Variable-Based Calibration for Machine Learning Classifiers [11.9995808096481]
  We introduce the notion of variable-based calibration to characterize calibration properties of a model. We find that models with near-perfect expected calibration error can exhibit significant miscalibration as a function of features of the data.
  arXiv  Detail & Related papers  (2022-09-30T00:49:31Z)
• Fair admission risk prediction with proportional multicalibration [0.16249424686052708]
  Multicalibration constrains calibration error among flexibly-defined subpopulations. It is possible for a decision-maker to learn to trust or distrust model predictions for specific groups. We propose proportional multicalibration, a criteria that constrains the percent calibration error among groups and within prediction bins.
  arXiv  Detail & Related papers  (2022-09-29T08:15:29Z)
• Unsupervised Calibration under Covariate Shift [92.02278658443166]
  We introduce the problem of calibration under domain shift and propose an importance sampling based approach to address it. We evaluate and discuss the efficacy of our method on both real-world datasets and synthetic datasets.
  arXiv  Detail & Related papers  (2020-06-29T21:50:07Z)

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.