mlsanityData to 7 Oct 2026

STAMP: Predicting Out-of-Distribution Generalization without Target Data

arXiv 2609.32672 ↗First posted 26 Sep 2026cs.LG cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Predicting whether a trained model will generalize under distribution shift remains difficult, especially when target-domain data are unavailable. We introduce STAMP (Semantic Temporal Augmented Model Prediction), a source-only, target-label-free criterion that estimates out-of-distribution (OOD) performance from paired source-domain images. STAMP computes the output-space correlation ratio η2=SB/ST by contrasting semantically stable pairs with random pairs: higher η2 indicates that model outputs vary with semantic identity rather than nuisance variation. On 44 chest X-ray models spanning CNNs, ViTs, MetaFormers, foundation models, and SSL/VLM probes, temporal STAMP attains Spearman correlations of 0.844--0.855 with macro AUROC on VinDr-CXR, CheXpert, and MIMIC-CXR; a class-matched variant improves single-class RSNA from 0.311 to 0.663. STAMP attains the best average source-only medical ranking and outperforms the target-domain ATC and AoTL estimators without any target data. On 27 ImageNet models, temperature-scaled STAMPTS attains ρ=0.984 on ObjectNet and ρ≥0.905 on four additional distribution shifts, with partial correlations of 0.662--0.949 after controlling for ImageNet accuracy. Requiring approximately 12s per model on one GPU, STAMP is a practical pre-deployment model-selection and auditing tool.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 36 of its 52 references that are in the map. First posted on arXiv 26 Sep 2026.

Builds on

36 of 52 references.

References · 36

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  3. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,082
  4. 2016Densely Connected Convolutional Networkscited by 6,691
  5. 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,415
  6. 2017On Calibration of Modern Neural Networkscited by 4,860
  7. 2020Training data-efficient image transformers & distillation through attentioncited by 4,560
  8. 2022A ConvNet for the 2020scited by 3,545
  9. 2019Benchmarking Neural Network Robustness to Common Corruptions and Perturbationscited by 2,857
  10. 2020Shortcut Learning in Deep Neural Networkscited by 1,928
  11. 2020The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalizationcited by 1,842
  12. 2021MLP-Mixer: An all-MLP Architecture for Visioncited by 1,441
24 more
  1. 2019CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparisoncited by 1,410
  2. 2019Natural Adversarial Examplescited by 1,401
  3. 2017ChestX-ray8: Hospital-scale Chest X-ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseasescited by 969
  4. 2021CoAtNet: Marrying Convolution and Attention for All Data Sizescited by 566
  5. 2021MViTv2: Improved Multiscale Vision Transformers for Classification and Detectioncited by 516
  6. 2022MaxViT: Multi-Axis Vision Transformercited by 476
  7. 2018Confounding variables can degrade generalization performance of radiological deep learning modelscited by 374
  8. 2021Accuracy on the Line: On the Strong Correlation Between Out-of-Distribution and In-Distribution Generalizationcited by 248
  9. 2020VinDr-CXR: An open dataset of chest X-rays with radiologist's annotationscited by 236
  10. 2020LEEP: A New Measure to Evaluate Transferability of Learned Representationscited by 182
  11. 2021LogME: Practical Assessment of Pre-trained Models for Transfer Learningcited by 168
  12. 2022Leveraging Unlabeled Data to Predict Out-of-Distribution Performancecited by 141
  13. 2021TorchXRayVision: A library of chest X-ray datasets and modelscited by 131
  14. 2022MetaFormer Baselines for Visioncited by 131
  15. 2024Exploring scalable medical image encoders beyond text supervisioncited by 129
  16. 2021Predicting with Confidence on Unseen Distributionscited by 116
  17. 2016Theoretical Analysis of Domain Adaptation with Optimal Transportcited by 105
  18. 2022An Information-Theoretic Approach to Transferability in Task Transfer Learningcited by 79
  19. 2021What Does Rotation Prediction Tell Us about Classifier Accuracy under Varying Testing Environments?cited by 63
  20. 2019Can we trust deep learning models diagnosis? The impact of domain shift in chest radiograph classificationcited by 57
  21. 2022Predicting Out-of-Distribution Error with the Projection Normcited by 50
  22. 2021CheXternal: Generalization of Deep Learning Models for Chest X-ray Interpretation to Photos of Chest X-rays and External Clinical Settingscited by 5
  23. 2026Representation Geometry as a Diagnostic for Out-of-Distribution Robustnesscited by 3
  24. 2026TopoGeoScore: A Self-Supervised Source-Only Geometric Framework for OOD Checkpoint Selectioncited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2022RoentGen: Vision-Language Foundation Model for Chest X-ray Generation
  2. 2025Delving into Out-of-Distribution Detection with Medical Vision-Language Models
  3. 2026A data- and compute-efficient chest X-ray foundation model beyond aggressive scaling
  4. 2025Open-PMC-18M: A High-Fidelity Large Scale Medical Dataset for Multimodal Representation Learning
  5. 2026Flow Matching with Optimized Subclass Priors for Medical Image Augmentation
  6. 2023UniChest: Conquer-and-Divide Pre-training for Multi-Source Chest X-Ray Classification
  7. 2026DobicVLM: Aligning Chest X-Ray Report Generation with Clinically-Grounded Programmatic Rewards via Group Relative Policy Optimization
  8. 2026Label-Free Foundational Model Selection for Medical Image Classification under Distribution Shift via Pseudo Label Discrepancy