The paper introduces CoFailMed, an audit framework that examines correlated failures among pretrained encoder pools used for medical image segmentation. By evaluating 11 public encoders across five 2D tasks with frozen features and 1×1 readouts, the study finds that models sharing checkpoints exhibit the highest per‑case Dice correlation, and that distinct checkpoints within the same encoder family are more correlated than cross‑family pairs. The analysis also shows that richer readouts and fine‑tuning can narrow but not eliminate these correlations, and that effective pool size at a 33rd‑percentile Dice threshold varies with encoder family and training recipe, underscoring the need to report absolute dependence and threshold‑specific effective size alongside accuracy.
By Eunseob Choi, Kyeonghun Kim, Hyuk-Jae Lee, Nam-Joon Kim
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.
By Mahmoud Ibrahim, Bart Elen, Chang Sun, Gokhan Ertaylan, Michel Dumontier
The paper evaluates how readout settings and operating points affect event-stream candidate selection in event-triggered recording. Using 520 DailyDVS‑200 clips and 47 participants, the authors compare several encoder configurations—including DINO‑CLS, patch mean, and MobileNetV3‑Small—across different accumulation durations and view counts, reporting metrics such as target capture, false triggers, and partial AUC. The study finds that lightweight baselines and specific operating points significantly influence performance, while no clear advantage is observed for the JEPA‑specific method.
By Yifan Shuai
arXiv:2606. 10066v1 Announce Type: cross Abstract: Medical vision-language models (VLMs) are evaluated on public benchmarks whose images and question-answer pairs have been freely downloadable for years, yet reported accuracy assumes these examples were absent from pretraining.
By Bruce Changlong Xu, Lan Wu, Alexander Ryu
arXiv:2610.01284v1 Announce Type: cross
Abstract: Model validation estimates the performance of a complete learning procedure on new data. However, an invalid split can produce an optimistic and stab...
By Mehmet Baygin, Sengul Dogan, Turker Tuncer
The paper investigates how to evaluate bounded model updates for ICU prediction models using the MIMIC‑IV database. After auditing the data for recording artifacts and cleaning it, the authors compare a bounded update that retrains only treatment‑context and fusion parameters with full fine‑tuning across two model families and two targets, finding that bounded updates are non‑inferior or superior in most cases but depend on target, model family, and learning‑rate tuning. The study also shows that audits should precede update evaluation and that the update boundary should be chosen per target.
By Fatema Ferdous Tamanna, K. M. Merajul Arefin, Md. Abdul Masud
arXiv:2607. 25589v1 Announce Type: cross Abstract: Medical-imaging AI benchmarks combine datasets, DICOM rendering, prompts, provider APIs, automated labels, statistical code, manuscripts, and repository releases.
By Mateusz Koz{\l}owski
arXiv:2606. 02276v1 Announce Type: cross Abstract: Vision-language models (VLMs) trained on paired chest radiographs and radiology reports learn a shared embedding space that can preserve instance-level image-report correspondence.
By Soroosh Tayebi Arasteh, Mahshad Lotfinia, Sven Nebelung, Daniel Truhn
arXiv:2609.37848v1 Announce Type: cross
Abstract: Biomedical machine learning papers often compress model performance into one headline number. That number can look like a property of the model even...
By Bhanu Prakash Vangala, Sowmya Guda, Latha Peddi, Navya Vangala
CropCop is a closed‑set plant‑health recognition system covering 120 operational classes, built from a rigorously audited dataset of 109,107 images after removing 3,233 duplicate relationships. The model, based on a fine‑tuned DINOv3 ConvNeXt‑Tiny, achieves 98.51% accuracy and 96.87% macro‑F1 on a locked internal test, while a quantised MobileNetV4 variant reaches 98.46% accuracy and 96.23% macro‑F1 in a 22.60 MiB runtime artifact. Validation‑only post‑training quantisation and a compact ExecuTorch/XNNPACK PTE ensure high fidelity between the trained model and its deployed form, with minimal decision changes between the INT8 graph and the final artifact.
By Rana Muhammad Ahmed, Sabahat Abbas
arXiv:2609. 31450v1 Announce Type: cross Abstract: Medical vision-language model (VLM) post-training is commonly evaluated through answer accuracy.
By Wang Jingxin
PairCLIP‑SWA is a single‑backbone deepfake detector that fine‑tunes an OpenCLIP ViT‑L/14 visual tower using a class‑balanced corpus created by CLIP‑similarity‑guided subsampling and then applies stochastic weight averaging over the fine‑tuning trajectory. Trained on the IEEE SP Cup 2025 (DF‑Wild‑Cup) benchmark, the weight‑averaged model achieves 97.04 % accuracy, 98.64 % AU‑ROC and a 4.07 % equal error rate—outperforming the best published three‑backbone ensemble. The study shows that averaging stabilizes validation performance, prevents misranking on disjoint test sets, and that CLIP pre‑training, rather than ensembling, is the key driver of accuracy.
By Md. Nakib, Sudipto Sarkar, A. J. A. Jubayer Talukder, Mohammad Ariful Haque