3 papers
cs.CV2026
ModaLens: Measuring Image Sensitivity in Report-Conditioned Medical VLMs
Sebastián Andrés Cajas Ordóñez, Maximin Lange, Quang Bui +9
A radiology report can already answer a clinical question, so it is hard to tell whether a vision-language model also uses the image. ModaLens, a paired image-swap audit, measures…
cs.AI2026
Agents Catching Agents: Shortcut Cascades and Benchmark Gaming in Clinical Multi-Agent Systems
Sebastián Andrés Cajas Ordóñez, Agastya Munnangi, Aldo Marzullo +13
Clinical decision support is moving toward committees of language-model agents deliberating on a shared workspace. We ask whether such committees can be gamed by shortcuts, cues a…
cs.AI2026
Loud or Silent? A Reusable Framework for Per-Modality Failure Analysis in Multimodal Clinical AI
Quang Bui, Shlok Jaiswal, Samuel Paik-Heintz +14
Multimodal clinical models are usually judged on accuracy with every modality present, but deployment removes modalities; an echocardiogram is often unavailable where an ECG is rou…