Beyond Discrete Tokens: Preserving Numerical Continuity in Medical Vision-Language Models

Dianye Huang, Jiamei Guo, Yue Zhou, Mingde Li, Yi Xiong et al.|October 23, 2026arXiv

Key Takeaway

Combining diffusion-based continuous prediction with language models outperforms discrete token prediction for medical measurements, suggesting hybrid architectures better match the continuous nature of real-world quantitative tasks.

Summary

This paper presents HAD-VLM, a medical AI system that combines language understanding with precise numerical measurement from ultrasound and X-ray images.

multimodal architecture

Key Terms

diffusion-process vision-language-models prompt-engineering visual-pretraining ensemble-prediction