Entity · paper

Hearsay: Vision-Language Medical Diagnoses Without an Image

paperactiveprovisionalhearsay-vision-language-medical-diagnoses-without-an-image-4f134685·1 events·first seen 44h ago

Aliases: Hearsay: Vision-Language Medical Diagnoses Without an Image

Co-occurring entities

Gemini 3.1 Pro Claude Opus 4.6 Google OpenAI GPT-5.5 Anthropic

More like this (12)

Vision-Language Models Small Vision-Language Models Know When They Are Wrong But Cannot Say So Connecting Speech to Words through Images Seeing Is Not Sharing: Some Vision-Language Models Overestimate Common Ground in Asymmetric Dialogue LabVLA: Grounding Vision-Language-Action Models in Scientific Laboratories vision-language grounding Vision-OPD Lesioned Multimodal Language Models Reproduce Aphasic Picture-Naming Patterns Vision-Default, Prior-Override: Causal Mechanisms of Perception-Knowledge Conflict in Vision-Language Models Vision-Language-Action model Vision-Language-Action models visual language model

Recent events (1)

8arXiv · cs.CL·44h ago·source ↗

Frontier VLMs confabulate demographic-biased diagnoses when no medical image is provided

A new arXiv paper demonstrates that Claude Opus 4.7, GPT-5.4, and Gemini 3.1 Pro will fabricate structured medical diagnoses when queried with only a patient demographic descriptor and no image attached, rather than abstaining. The confabulation is systematically biased by patient demographics — e.g., Sarcoidosis is disproportionately diagnosed for young Black patients on chest X-ray prompts. The paper identifies a critical dissociation where prose output hedges about the missing image while the structured diagnosis field still names a disease, making the failure invisible to prose-only audits, and shows that the effect is sensitive to specific probe words, suggesting multiple distinct failure modes.

Evaluation and Benchmarking AI Safety Research Gemini 3.1 Pro Claude Opus 4.6 Hearsay: Vision-Language Medical Diagnoses Without an Image +5 more