Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2606.01543 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910279273545728 |
|---|---|
| author | Zhang, Yuan Xia, Jiahao Huang, Junzhang Wang, Meng Chen, Feng Yang, Guanyu Fu, Huazhu |
| author_facet | Zhang, Yuan Xia, Jiahao Huang, Junzhang Wang, Meng Chen, Feng Yang, Guanyu Fu, Huazhu |
| contents | Data scarcity in multimodal pathology motivates unified generative models that synthesize modality-specific appearance while preserving anatomically coherent structure. Although modalities differ in appearance statistics, morphological structures such as cellular topology and tissue boundaries are largely preserved across acquisition protocols. However, existing methods often model these factors within a homogeneous token stream, implicitly coupling structure with appearance and weakening structural controllability under modality shifts. To address this, we propose pathology Autorgressive modeling (PathAR), a structure-first autoregressive synthesis framework that explicitly factorizes structure and appearance for modality-label-conditioned pathology generation.PathAR employs a dual vector quantization (Dual-VQ) tokenizer to decompose samples into mask-grounded structure and appearance tokens, and an interleaved autoregressive (IAR) transformer with asymmetric attention visibility to enforce structure-to-appearance dependence. PathAR stabilizes morphology under heterogeneous modality-specific appearances and enables spatially aligned image--mask pair generation. Extensive experiments show that PathAR improves structural consistency and modality fidelity over baselines, maintains sample diversity, supports downstream segmentation in data-scarce regimes, and demonstrates extensibility to finer-grained intra-modality organ-label variation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2606_01543 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | PathAR: Structure-First Autoregressive Synthesis of Multimodal Pathology Images Zhang, Yuan Xia, Jiahao Huang, Junzhang Wang, Meng Chen, Feng Yang, Guanyu Fu, Huazhu Computer Vision and Pattern Recognition Data scarcity in multimodal pathology motivates unified generative models that synthesize modality-specific appearance while preserving anatomically coherent structure. Although modalities differ in appearance statistics, morphological structures such as cellular topology and tissue boundaries are largely preserved across acquisition protocols. However, existing methods often model these factors within a homogeneous token stream, implicitly coupling structure with appearance and weakening structural controllability under modality shifts. To address this, we propose pathology Autorgressive modeling (PathAR), a structure-first autoregressive synthesis framework that explicitly factorizes structure and appearance for modality-label-conditioned pathology generation.PathAR employs a dual vector quantization (Dual-VQ) tokenizer to decompose samples into mask-grounded structure and appearance tokens, and an interleaved autoregressive (IAR) transformer with asymmetric attention visibility to enforce structure-to-appearance dependence. PathAR stabilizes morphology under heterogeneous modality-specific appearances and enables spatially aligned image--mask pair generation. Extensive experiments show that PathAR improves structural consistency and modality fidelity over baselines, maintains sample diversity, supports downstream segmentation in data-scarce regimes, and demonstrates extensibility to finer-grained intra-modality organ-label variation. |
| title | PathAR: Structure-First Autoregressive Synthesis of Multimodal Pathology Images |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2606.01543 |