Multimodal diffusion framework for collaborative text image audio generation and applications
Abstract This paper presents a novel framework for collaborative generation across text, image, and audio modalities using an enhanced diffusion model architecture. We introduce a Hierarchical Cross-modal Alignment Network that establishes unified representations while preserving modality-specific c...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-07-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-05794-4 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|