Disentangled Representation Learning via Flow Matching
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866917478708281344 |
|---|---|
| author | Chi, Jinjin Liu, Taoping Yin, Mengtao Li, Ximing Jing, Yongcheng Shen, Jialie Rutkowski, Leszek Tao, Dacheng |
| author_facet | Chi, Jinjin Liu, Taoping Yin, Mengtao Li, Ximing Jing, Yongcheng Shen, Jialie Rutkowski, Leszek Tao, Dacheng |
| contents | Disentangled representation learning aims to capture the underlying explanatory factors of observed data, enabling a principled understanding of the data-generating process. Recent advances in generative modeling have introduced new paradigms for learning such representations. However, existing diffusion-based methods encourage factor independence via inductive biases, yet frequently lack strong semantic alignment. In this work, we propose a flow matching-based framework for disentangled representation learning, which casts disentanglement as learning factor-conditioned flows in a compact latent space. To enforce explicit semantic alignment, we introduce a non-overlap (orthogonality) regularizer that suppresses cross-factor interference and reduces information leakage between factors. Extensive experiments across multiple datasets demonstrate consistent improvements over representative baselines, yielding higher disentanglement scores as well as improved controllability and sample fidelity. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_05214 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Disentangled Representation Learning via Flow Matching Chi, Jinjin Liu, Taoping Yin, Mengtao Li, Ximing Jing, Yongcheng Shen, Jialie Rutkowski, Leszek Tao, Dacheng Machine Learning Disentangled representation learning aims to capture the underlying explanatory factors of observed data, enabling a principled understanding of the data-generating process. Recent advances in generative modeling have introduced new paradigms for learning such representations. However, existing diffusion-based methods encourage factor independence via inductive biases, yet frequently lack strong semantic alignment. In this work, we propose a flow matching-based framework for disentangled representation learning, which casts disentanglement as learning factor-conditioned flows in a compact latent space. To enforce explicit semantic alignment, we introduce a non-overlap (orthogonality) regularizer that suppresses cross-factor interference and reduces information leakage between factors. Extensive experiments across multiple datasets demonstrate consistent improvements over representative baselines, yielding higher disentanglement scores as well as improved controllability and sample fidelity. |
| title | Disentangled Representation Learning via Flow Matching |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2602.05214 |