Shortcut learning occurs when classifiers rely on spurious correlations instead of true diagnostic features. The authors argue that this is essentially a calibration issue: unconstrained training aligns shortcut groups with their prevalence, causing over‑confidence in one group and under‑confidence in another. They introduce two encoder‑agnostic methods—a regularizer and a post‑hoc recalibration—to equalize prevalence across shortcut groups, achieving significant performance gains on chest‑drain‑pneumothorax benchmarks with both fine‑tuned CNNs and frozen foundation‑model backbones.
arXiv:2608. 12086v1 Announce Type: cross Abstract: Vision-language models, such as contrastive language-image pre-training (CLIP)-based approaches, have reached state-of-the-art (SOTA) results in medical artificial intelligence.
MVC-Bench is a new benchmark designed to evaluate the calibration of vision‑language models (VLMs) and medical VLMs (Medical‑VLMs) for medical image classification. It tests calibration across robustness to modality, backbone, and domain shift; effectiveness of calibration strategies and prompt‑tuning methods; and stability under prompt‑template and random‑seed variations. The benchmark includes eight backbones, three medical modalities (fundus imaging, histopathology, chest X‑ray), and compares post‑hoc, train‑time, and zero‑shot calibration approaches, reporting accuracy, Expected Calibration Error (ECE), Maximum Calibration Error (MCE), and Adaptive Calibration Error (ACE) over 1,638 experiments, while also proposing a Multi‑Class Margin (MCM) regularization technique that improves ECE in most settings.
arXiv:2605. 30188v2 Announce Type: replace-cross Abstract: Reliable probability estimates are critical in many machine learning applications, yet modern classifiers are often poorly calibrated.
arXiv:2607. 14984v1 Announce Type: new Abstract: Per-subgroup fairness audits of medical image classifiers face a sample-size problem: minority subgroups in held-out test sets have so few samples that the resulting confidence intervals on per-subgroup performance are wider than the bias the audit is meant to detect.
The study compares classical and deep learning classifiers for a five‑class respiratory‑sound dataset of 1,211 clips. An SVM using 30 summary features (MFCC, ZCR, RMS) achieves macro recall of 0.78, while a 1.13M‑parameter convolutional‑recurrent network reaches 0.75–0.77; no clear advantage is found between model families. The largest performance gain comes from standardizing the SVM’s features, which boosts macro recall by 27 points, far exceeding any architectural difference.
MVC-Bench is a calibration-focused benchmark for medical vision‑language models, evaluating how well these models express confidence across different modalities, backbones, and domain shifts. It tests robustness to modality, backbone, and domain changes, the effectiveness of calibration and prompt‑tuning strategies, and stability under prompt‑template and random‑seed variations. The benchmark includes 1638 experiments, reporting accuracy and Expected Calibration Error (ECE) along with other calibration metrics, and introduces a simple train‑time calibration method, Multi‑Class Margin (MCM) regularization, that achieves the lowest ECE in most settings.
arXiv:2606. 28654v1 Announce Type: cross Abstract: Deep Neural Network (DNN) classifiers suffer from poor calibration when their softmax outputs (predictive confidence) deviate from the empirical likelihoods.
arXiv:2608.29419v1 Announce Type: cross Abstract: Deep learning architectures are increasingly proposed for patient trajectory modeling in electronic health records (EHRs), yet their advantage over s...
GRIN+ is a new machine unlearning framework that targets fast and precise data erasure in imbalanced medical datasets. It separates unlearning‑specific knowledge from general representations by analyzing gradient contributions of forget and retain sets, introduces a class‑adaptive influence scoring to counter gradient dominance, and uses a direction‑constrained update to protect essential clinical knowledge. Benchmarks on skin cancer, brain tumor, and breast ultrasound data show that GRIN+ balances privacy, efficiency, and utility, achieving high diagnostic accuracy and faster runtime than existing methods.
arXiv:2509. 19671v3 Announce Type: replace Abstract: Public datasets of Chest X-Rays (CXRs) have long been a popular benchmark for developing machine learning (ML) computer vision models in healthcare.
The paper evaluates three architectures for multi-label retinal disease classification on a unified 39‑target dataset. A staged gated cascade, a flat sigmoid model, and a label‑correlation transformer are compared, all sharing the same backbone and training protocol. The label‑correlation model yields the largest performance gain, and blending it with the flat model under test‑time augmentation achieves the best results across most metrics.
The paper presents DualTTA, a model‑agnostic framework that improves the calibration of black‑box radiology AI systems by applying clinically grounded test‑time augmentations (geometric and physics‑inspired 3D CT perturbations) and learning probability‑level aggregation strategies. Without accessing model internals or training data, DualTTA achieved the best overall calibration across pulmonary embolism and intracranial hemorrhage detection tasks, reducing Expected Calibration Error by 54% and 43% respectively. It also outperformed traditional uncertainty estimation methods that require internal model access, such as Temperature Scaling, MC Dropout, and Deep Ensembles.