DiEC: Diffusion Embedded Clustering
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866915742031544320 |
|---|---|
| author | Hu, Haidong Zheng, Xiaoyu Zhou, Jin Wang, Yingxu Wang, Rui Dong, Pei Han, Shiyuan Wang, Lin Chen, C. L. Philip Zhang, Tong Chen, Yuehui |
| author_facet | Hu, Haidong Zheng, Xiaoyu Zhou, Jin Wang, Yingxu Wang, Rui Dong, Pei Han, Shiyuan Wang, Lin Chen, C. L. Philip Zhang, Tong Chen, Yuehui |
| contents | Deep clustering methods typically rely on a single, well-defined representation for clustering. In contrast, pretrained diffusion models provide abundant and diverse multi-scale representations across network layers and noise timesteps. However, a key challenge is how to efficiently identify the most clustering-friendly representation in the layer*timestep space. To address this issue, we propose Diffusion Embedded Clustering (DiEC), an unsupervised framework that performs clustering by leveraging optimal intermediate representations from pretrained diffusion models. DiEC systematically evaluates the clusterability of representations along the trajectory of network depth and noise timesteps. Meanwhile, an unsupervised search strategy is designed for recognizing the Clustering-optimal Layer (COL) and Clustering-optimal Timestep (COT) in the layer*timestep space of pretrained diffusion models, aiming to promote clustering performance and reduce computational overhead. DiEC is fine-tuned primarily with a structure-preserving DEC-style KL-divergence objective at the fixed COL + COT, together with a random-timestep diffusion denoising objective to maintain the generative capability of the pretrained model. Without relying on augmentation-based consistency constraints or contrastive learning, DiEC achieves excellent clustering performance across multiple benchmark datasets. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_20905 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | DiEC: Diffusion Embedded Clustering Hu, Haidong Zheng, Xiaoyu Zhou, Jin Wang, Yingxu Wang, Rui Dong, Pei Han, Shiyuan Wang, Lin Chen, C. L. Philip Zhang, Tong Chen, Yuehui Machine Learning Artificial Intelligence Deep clustering methods typically rely on a single, well-defined representation for clustering. In contrast, pretrained diffusion models provide abundant and diverse multi-scale representations across network layers and noise timesteps. However, a key challenge is how to efficiently identify the most clustering-friendly representation in the layer*timestep space. To address this issue, we propose Diffusion Embedded Clustering (DiEC), an unsupervised framework that performs clustering by leveraging optimal intermediate representations from pretrained diffusion models. DiEC systematically evaluates the clusterability of representations along the trajectory of network depth and noise timesteps. Meanwhile, an unsupervised search strategy is designed for recognizing the Clustering-optimal Layer (COL) and Clustering-optimal Timestep (COT) in the layer*timestep space of pretrained diffusion models, aiming to promote clustering performance and reduce computational overhead. DiEC is fine-tuned primarily with a structure-preserving DEC-style KL-divergence objective at the fixed COL + COT, together with a random-timestep diffusion denoising objective to maintain the generative capability of the pretrained model. Without relying on augmentation-based consistency constraints or contrastive learning, DiEC achieves excellent clustering performance across multiple benchmark datasets. |
| title | DiEC: Diffusion Embedded Clustering |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2512.20905 |