Distribution Shift

Ask about this topic

Overview

Distribution shift occurs when the data a model encounters in deployment differs from its training data, through changes in inputs, label frequencies, relationships, domains, or time. Researchers study detection and measurement with density-ratio methods, statistical tests, and representation-based diagnostics, and seek to reduce failures through domain generalization, importance weighting, calibration, and test-time adaptation. Evaluation spans settings such as medical imaging across hospitals, temporal forecasting, and vision under changes in weather or sensors, where random splits can conceal deployment risks. A central tension is that adaptation can improve performance but also amplify errors when shift estimates are unreliable or target labels are scarce, making dependable monitoring and realistic evaluation essential.

Latest papers

CardsList

179 papers24 papers in the last four weeks, level with the four weeks before. 0.1% of all new papers.

  1. Development and evaluation of CADe systems in low-prevalence setting: The RARE25 challenge for early detection of Barrett's neoplasia

    Apr 13, 2026Tim J. M. Jaspers, Francisco Caetano, Cris H. B. Claessens +8Medical Image AnalysisMedical Image Benchmarks

    Department of Electrical Engineering, Architectures for Reliable Image Analysis Lab, Eindhoven University of Technology, Eindhoven, Netherlands · Department of Gastroenterology and Hepatology, Amsterdam University Medical Centers, University of Amsterdam, Amsterdam, The Netherlands

  2. Generalization Under Scrutiny: Cross-Domain Detection Progresses, Pitfalls, and Persistent Challenges

    Apr 9, 2026Saniya M. Deshmukh, Kailash A. Hambarde, Hugo ProençaObject DetectionCross-Domain Generalization

    IT: Instituto de Telecomunicações, University of Beira Interior, Covilhã, Portugal.

  3. FedDAP: Domain-Aware Prototype Learning for Federated Learning under Domain Shift

    Apr 8, 2026Huy Q. Le, Loc X. Nguyen, Yu Qiao +3Federated LearningDistribution Shift

    G-LAMP NEXUS Institute, Kyung Hee University · Kyung Hee University

  4. Investigating Data Interventions for Subgroup Fairness: An ICU Case Study

    Apr 3, 2026Erin Tan, Judy Hanwen Shen, Irene Y. ChenGroup FairnessTraining Data Selection

    University of California, Berkeley · Stanford University

  5. Analytic Drift Resister for Non-Exemplar Continual Graph Learning

    Apr 3, 2026Lei Song, Shihan Guan, Youyong KongIncremental LearningClass-Incremental Learning

    School of Computer Science and Engineering, Southeast University, Nanjnig 211189, China

  6. Reasoning as Energy Minimization over Structured Latent Trajectories

    Mar 30, 2026David K. JohanssonLatent-Space PlanningContinuous Latent Reasoning

    Polished Snow Inc.

  7. Koopman-based surrogate modeling for reinforcement-learning-control of Rayleigh-Benard convection

    Mar 30, 2026Tim Plotzki, Sebastian PeitzRL ControlPDE Surrogate Modeling

    TU Dortmund University & Lamarr Institute for Machine Learning and Artificial Intelligence, Dortmund, Germany

  8. Edge Radar Material Classification Under Geometry Shifts

    Mar 24, 2026Jannik Hohmann, Dong Wang, Andreas NüchtermmWave Radar SensingDistribution Shift

    Department of Informatics XVII (Robotics), Julius-Maximilians-University Würzburg, 97074 Würzburg, Germany. · Zentrum für Telematik e.V., Würzburg · U2IS, ENSTA, Institut Polytechnique de Paris, France.

  9. How to Fine-Tune a Reasoning Model? A Teacher-Student Cooperation Framework to Synthesize Student-Consistent SFT Data

    Mar 23, 2026Zixian Huang, Kaichen Yang, Xu Huang +6Synthetic Data Generation for Language ModelsSupervised Fine-Tuning

    Shanghai AI Laboratory · Dalian University of Technology · Nanjing University

  10. Dual-level Adaptation for Multi-Object Tracking: Building Test-Time Calibration from Experience and Intuition

    Mar 23, 2026Wen Guo, Pengfei Zhao, Zongmeng Wang +2Multi-Object TrackingTest-Time Adaptation

    Shandong Technology and Business University · Inner Mongolia University · University of Science and Technology Beijing +1

  11. DirPA: Addressing Prior Shift in Imbalanced Few-shot Crop-type Classification

    Mar 13, 2026Joana Reuss, Ekaterina Gikalo, Marco KörnerClass-Imbalanced LearningDistribution Shift

    Technical University of Munich (TUM), TUM School of Engineering and Design, Department of Aerospace and Geodesy, Chair of Remote Sensing Technology, Munich, 80333, Germany · Technical University of Munich (TUM), Munich Data Science Institute (MDSI), Garching, 85748, Germany · ELLIS Unit Jena, University of Jena, Jena, Germany, Jena, 07743, Germany

  12. Beyond the Class Subspace: Teacher-Guided Training for Reliable Out-of-Distribution Detection in Single-Domain Models

    Mar 11, 2026Hong Yang, Devroop Kar, Qi Yu +2OOD DetectionDistribution Shift

    Rochester Institute of Technology, Rochester, NY 14623, USA

  13. The Confidence Gate Theorem: When Should Ranked Decision Systems Abstain?

    Mar 10, 2026Ronald DokuDecision-Making under UncertaintyUncertainty Quantification

    Haske Labs

  14. Impermanent: A Live Benchmark for Temporal Generalization in Time Series Forecasting

    Mar 9, 2026Azul Garza, Renée Rosillo, Rodrigo Mendoza-Smith +5Forecasting BenchmarksTime Series Forecasting

    TimeCopilot · Independent Researcher · ELLIS Institute Tübingen +3

  15. Drift-to-Action Controllers: Budgeted Interventions with Online Risk Certificates

    Mar 9, 2026Ismail Lamaakal, Chaymae Yahyati, Khalid El Makkaoui +2Drifting ModelsDistribution Shift

    Multidisciplinary Faculty of Nador, Mohammed First University, Oujda 60000, Morocco · Laboratory LaSTI, ENSAK, Sultan Moulay Slimane University, Khouribga 54000, Morocco

  16. An Efficient and Effective Evaluator for Text2SQL Models on Unseen and Unlabeled Data

    Mar 8, 2026Trinh Pham, Thanh Tam Nguyen, Viet Huynh +2Automated EvaluationText-to-SQL

    Griffith University (Australia) · Edith Cowan University (Australia) · The University of Queensland (Australia)

  17. ConfHit: Conformal Generative Design with Oracle Free Guarantees

    Mar 7, 2026Siddhartha Laghuvarapu, Ying Jin, Jimeng SunMolecular GenerationWeighted Conformal Prediction

    Siebel School of Computing and Data Science, University of Illinois Urbana-Champaign, IL, USA · Department of Statistics and Data Science, Wharton School, University of Pennysylvania, PA, USA

  18. Towards High-resolution and Disentangled Reference-based Sketch Colorization

    Mar 6, 2026Dingkun Yan, Xinrui Wang, Ru Wang +5Distribution ShiftImage-to-Image Translation

    The University of Tokyo · Project HAT · Waseda University

  19. Post Fusion Bird's Eye View Feature Stabilization for Robust Multimodal 3D Detection

    Mar 5, 2026Trung Tien Dong, Dev Thakkar, Arman Sargolzaei +1Multimodal Robustness3D Object Detection

    College of Engineering, University of South Florida, Tampa, FL 33620, USA

  20. On the Learnability of Offline Model-Based Optimization: A Ranking Perspective

    Mar 4, 2026Shen-Huan Lyu, Rong-Xi Tan, Ke Xue +4Surrogate ModelingDistribution Shift

    Key Laboratory of Water Big Data Technology of Ministry of Water Resources, College of Computer Science and Software Engineering, Hohai University, Nanjing, China · Department of Computer Science, City University of Hong Kong, Hong Kong, China · National Key Laboratory for Novel Software Technology, Nanjing University, Nanjing, China +1

  21. I Can't Believe It's Not Robust: Catastrophic Collapse of Safety Classifiers under Embedding Drift

    Mar 1, 2026Subramanyam Sahoo, Vinija Jain, Divya Chaudhary +1Language Model RobustnessLLM Safety Evaluation

    Independent · Meta AI · Stanford University +2

  22. Fed-ADE: Adaptive Learning Rate for Federated Post-adaptation under Distribution Shift

    Mar 1, 2026Heewon Park, Mugon Joe, Miru Kim +2Adaptive Learning RateFederated Learning

    Department of Electrical and Computer Engineering, Sungkyunkwan University, Republic of Korea · Soongsil University, Seoul, Republic of Korea

  23. When Pretty Isn't Useful: Investigating Why Modern Text-to-Image Models Fail as Reliable Training Data Generators

    Feb 23, 2026Krzysztof Adamkiewicz, Brian Bernhard Moser, Stanislav Frolov +3Synthetic-to-Real Domain AdaptationDistribution Shift

    1RPTU University Kaiserslautern-Landau, Kaiserslautern, Germany · 2German Research Center for Artificial Intelligence (DFKI), Kaiserslautern, Germany

  24. Stabilizing Test-Time Adaptation of High-Dimensional Simulation Surrogates via D-Optimal Statistics

    Feb 17, 2026Anna Zimmel, Paul Setinek, Gianluca Galletti +2Test-Time AdaptationOOD Generalization

    ELLIS Unit, LIT AI Lab, Institute for Machine Learning, JKU Linz, Austria · Emmi AI GmbH, Linz, Austria

  25. TASTE: Task-Aware Out-of-Distribution Detection via Stein Operators

    Feb 7, 2026Michał Kozyra, Gesine ReinertDistribution Shift DetectionOOD Detection

    Department of Statistics, University of Oxford, United Kingdom.

  26. Factorizable joint shift revisited

    Jan 21, 2026Dirk TascheDistribution Shift RobustnessLabel Shift

    Centre for Business Mathematics and Informatics, North-West University, South Africa

  27. How to Teach Large Multimodal Models New Skills

    Oct 9, 2025Zhen Zhu, Yiming Gong, Yao Xiao +2Multimodal Large Language ModelsContinual Learning for VLMs

    University of Illinois Urbana-Champaign · Google DeepMind · Carnegie Mellon University

  28. Revisiting Performance Claims for Chest X-Ray Models Using Clinical Context

    Sep 24, 2025Andrew Wang, Jiashuo Zhang, Michael OberstChest X-Ray ClassificationDistribution Shift

    Brown University · Johns Hopkins University

  29. Analysis of On-policy Policy Gradient Methods under the Distribution Mismatch

    Mar 28, 2025Weizhen Wang, Xinzhe Zhou, Jianping He +1Policy GradientPolicy Gradient Methods

    School of Automation and Intelligent Sensing, Shanghai Jiao Tong University · Key Laboratory of System Control and Information Processing, Ministry of Education of China, Shanghai, China