.png){kind=icon}
.png){kind=icon}
Return
Redefining medical visual question answering using conditional generative diffusion models
.png){kind=icon}
PRE
AI
DOI：10.1016/j.bspc.2025.108222.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• This is the first work to apply diffusion models to medical visual question answering. • A Gaussian noising method embeds conditional info without breaking the Markov process. • The reverse process of the diffusion model reveals how answers are generated. • DiffuVQA shows strong reasoning and free-form answering ability in complex tasks.
Keywords:
diffusion models
medical visual question answering
Gaussian noising
Markov process
reasoning ability
Journal
IF:
4.9
Papers:
9.7K
Citations:
2.4W