analiza aiqbench
This research exposes a critical failure mode in vision-language models where they appear to utilize ECG data but fail to derive meaningful predictive value from specific patient records. The authors demonstrate that targeted visual prompt tuning can effectively mitigate this "ECG Mirage," significantly improving the model's ability to distinguish between matched and mismatched cardiac inputs for clinical deterioration prediction.
ECG Mirage: Revealing and Mitigating the Underutilisation of ECGs in Vision-Language Models for Clinical Prediction
arXiv:2609.21755v1 Announce Type: new Abstract: Emergency department (ED) decision-making relies on heterogeneous clinical information, including patient history, vital signs, laboratory results, and electrocardiograms (ECGs). Vision--language models (VLMs) can jointly process these modalities, but strong predictive performance does not necessarily imply meaningful use of the correct patient's ECG. We term this failure mode ECG Mirage: apparent multimodal capability without useful dependence on patient-specific ECG information. We distinguish two forms: ECG neglect, where ECGs provide little predictive benefit, and ECG confusion, where matched ECGs outperform no-image inputs but not mismatched ECGs. To evaluate these behaviours, we compare predictions obtained with matched ECGs, outcome-discordant mismatched ECGs, and no-image inputs while holding the clinical text and prediction targets fixed. Across four VLMs on MDS-ED, matched ECGs provide no consistent advantage for either ICU admission or clinical deterioration prediction. We then train four restricted visual prompts using supervised learning followed by conditional direct preference optimisation, while keeping the VLM backbone frozen. The resulting models achieve balanced accuracies of 70.6% for ICU admission and 67.5% for deterioration and increase the matched-versus-mismatched performance gap to approximately 16.5 and 5.5 percentage points, respectively. Overall, our study identifies ECG Mirage in multimodal clinical prediction and introduces visual prompt tuning as an efficient mitigation strategy.
Abstract:Emergency department (ED) decision-making relies on heterogeneous clinical information, including patient history, vital signs, laboratory results, and electrocardiograms (ECGs). Vision--language models (VLMs) can jointly process these modalities, but strong predictive performance does not necessarily imply meaningful use of the correct patient's ECG. We term this failure mode ECG Mirage: apparent multimodal capability without useful dependence on patient-specific ECG information. We distinguish two forms: ECG neglect, where ECGs provide little predictive benefit, and ECG confusion, where matched ECGs outperform no-image inputs but not mismatched ECGs. To evaluate these behaviours, we compare predictions obtained with matched ECGs, outcome-discordant mismatched ECGs, and no-image inputs while holding the clinical text and prediction targets fixed. Across four VLMs on MDS-ED, matched ECGs provide no consistent advantage for either ICU admission or clinical deterioration prediction. We then train four restricted visual prompts using supervised learning followed by conditional direct preference optimisation, while keeping the VLM backbone frozen. The resulting models achieve balanced accuracies of 70.6% for ICU admission and 67.5% for deterioration and increase the matched-versus-mismatched performance gap to approximately 16.5 and 5.5 percentage points, respectively. Overall, our study identifies ECG Mirage in multimodal clinical prediction and introduces visual prompt tuning as an efficient mitigation strategy.
Artificial Intelligence (cs.AI)
arXiv:2609.21755 [cs.AI]
(or arXiv:2609.21755v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2609.21755 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 294 cuvinte
Subiecte conexe
[AINews] not much happened today
Latent Space · 18 sept. 2026
The Mirage of Performance Gains: Why Contrastive Decoding Fails to Mitigate Object Hallucinations in MLLMs?
arXiv · 6 iun. 2026
MARCUS: An agentic, multimodal vision-language model for cardiac diagnosis and management
arXiv · 16 sept. 2026
Tri-PvP: Exposing Modality Bias in Omni-Modal Large Language Models through Perceptual-Propositional Evidence Conflicts
arXiv · 9 sept. 2026