Cover Image

6 Fingers, 1 Kidney: Natural Adversarial Medical Images Reveal Critical Weaknesses of Vision-Language Models

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Mayer, Leon, Kalinowski, Piotr, Ebersbach, Caroline, Knopp, Marcel, Rädsch, Tim, Christodoulou, Evangelia, Reinke, Annika, Kolbinger, Fiona R., Maier-Hein, Lena
Format:	Preprint
Published:	2025
Subjects:	Computer Vision and Pattern Recognition
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Challenging Vision-Language Models with Surgical Data: A New Dataset and Broad Benchmarking Study
by: Mayer, Leon, et al.
Published: (2025)

Bridging vision language model (VLM) evaluation gaps with a framework for scalable and cost-effective benchmark generation
by: Rädsch, Tim, et al.
Published: (2025)

Advancing standards in biomedical image analysis validation: A perspective on Metrics Reloaded
by: Annika Reinke, et al.
Published: (2025)

Quality Assured: Rethinking Annotation Strategies in Imaging AI
by: Rädsch, Tim, et al.
Published: (2024)

Beyond Knowledge Silos: Task Fingerprinting for Democratization of Medical Imaging AI
by: Godau, Patrick, et al.
Published: (2024)

False Promises in Medical Imaging AI? Assessing Validity of Outperformance Claims
by: Christodoulou, Evangelia, et al.
Published: (2025)

Performance uncertainty in medical image analysis: a large-scale investigation of confidence intervals
by: André, Pascaline, et al.
Published: (2026)

Confidence intervals uncovered: Are we ready for real-world medical imaging AI?
by: Christodoulou, Evangelia, et al.
Published: (2024)

FISBe: A real-world benchmark dataset for instance segmentation of long-range thin filamentous structures
by: Mais, Lisa, et al.
Published: (2024)

SURE-VQA: Systematic Understanding of Robustness Evaluation in Medical VQA Tasks
by: Kahl, Kim-Celine, et al.
Published: (2024)

Scientific approaches to technological officiating aids in game sports
by: Otto Kolbinger
Published: (2017)

Application-driven Validation of Posteriors in Inverse Problems
by: Adler, Tim J., et al.
Published: (2023)

Strategies to Improve Real-World Applicability of Laparoscopic Anatomy Segmentation Models
by: Kolbinger, Fiona R., et al.
Published: (2024)

Medical Imaging AI Competitions Lack Fairness
by: Reinke, Annika, et al.
Published: (2025)

Federated Learning for Surgical Vision in Appendicitis Classification: Results of the FedSurg EndoVis 2024 Challenge
by: Kirchner, Max, et al.
Published: (2025)

Field strength-dependent performance variability in deep learning-based analysis of magnetic resonance imaging
by: Qadir, Muhammad Ibtsaam, et al.
Published: (2025)

Revisiting 3D Medical Scribble Supervision: Benchmarking Beyond Cardiac Segmentation
by: Gotkowski, Karol, et al.
Published: (2024)

Adaptive-CaRe: Adaptive Causal Regularization for Robust Outcome Prediction
by: Bhasker, Nithya, et al.
Published: (2026)

Data Augmentation for Surgical Scene Segmentation with Anatomy-Aware Diffusion Models
by: Venkatesh, Danush Kumar, et al.
Published: (2024)

Refractory Motor Complications: Towards a Pragmatic Definition
by: Georg Ebersbach, et al.
Published: (2026)

Overcoming Common Flaws in the Evaluation of Selective Classification Systems
by: Traub, Jeremias, et al.
Published: (2024)

AI‐driven preoperative risk assessment in kidney cancer surgery: A comparative feasibility study of machine learning models
by: Julia Mühlbauer, et al.
Published: (2025)

Praxisform Polizeinotruf
by: Knopp, Philipp
Published: (2026)

Theory of functions / Konrad Knopp ; translated by Frederick Bagemihl
by: Knopp, Konrad

Station list and links to master tracks in different resolutions of POLARSTERN cruise ANT-XVI/4, Cape Town - Bremerhaven, 1999-05-11 - 1999-06-03
by: Reinke, Manfred
Published: (2015)

Station list and links to master tracks in different resolutions of POLARSTERN cruise ANT-XIX/1, Bremerhaven - Cape Town, 2001-11-07 - 2001-11-30
by: Reinke, Manfred
Published: (2015)

One model to use them all: Training a segmentation model with complementary datasets
by: Jenke, Alexander C., et al.
Published: (2024)

On Global Weak Solutions for the Magnetic Two-Component Hunter-Saxton System
by: Maier, Levin
Published: (2026)

Digital Literature and Critical Theory
by: Elstermann, Annika
Published: (2023)

Super Phantoms: advanced models for testing medical imaging technologies
by: Manohar, Srirang, et al.
Published: (2023)

Exploring Semantic Consistency in Unpaired Image Translation to Generate Data for Surgical Applications
by: Venkatesh, Danush Kumar, et al.
Published: (2023)

TwinTrack: Post-hoc Multi-Rater Calibration for Medical Image Segmentation
by: Kirscher, Tristan, et al.
Published: (2026)

Hybrid spectral-spatial domain registration for nanometric tracking in digital in-line holographic microscopy
by: Kalinowski, Kamil, et al.
Published: (2026)

The Moravian Graveyards of Lititz, Pa., 1744–1905
by: Beck, Abraham Reinke
Published: (2025)

QUANDO AS ROUPAS FALAM: DEBATE SOBRE A MODA COMO UMA FORMA DE LINGUAGEM
by: Carlos Augusto Reinke
Published: (2017)

A FUNÇÃO DO IMAGINÁRIO NO SIONISMO CRISTÃO BRASILEIRO
by: André Daniel Reinke
Published: (2023)

REPLY : Foslevodopa/Foscarbidopa Continuous Subcutaneous Infusion in Parkinson's Disease: Extended Real World Data With Outpatient Dose Titration
by: Katarina Rukavina, et al.
Published: (2025)

Making Web Applications GDPR Compliant: A Comparative Evaluation of GDPR-Enforcement Frameworks Artifact Evaluation
by: Felix Kalinowski
Published: (2025)

Calibrated Persistent Homology Tests for High-dimensional Collapse Detection
by: Kalinowski, Alexander
Published: (2026)

Monitoring Neural Training with Topology: A Footprint-Predictable Collapse Index
by: Kalinowski, Alexander
Published: (2026)