Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2603.11795 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911579700723712 |
|---|---|
| author | Shi, Hanyu Tao, Hong Huang, Guoheng Jiang, Jianbin Chen, Xuhang Pun, Chi-Man Wang, Shanhu Pan, Pan |
| author_facet | Shi, Hanyu Tao, Hong Huang, Guoheng Jiang, Jianbin Chen, Xuhang Pun, Chi-Man Wang, Shanhu Pan, Pan |
| contents | Unsupervised Concept Extraction aims to extract concepts from a single image; however, existing methods suffer from the inability to extract composable intrinsic concepts. To address this, this paper introduces a new task called Compositional and Interpretable Intrinsic Concept Extraction (CI-ICE). The CI-ICE task aims to leverage diffusion-based text-to-image models to extract composable object-level and attribute-level concepts from a single image, such that the original concept can be reconstructed through the combination of these concepts. To achieve this goal, we propose a method called HyperExpress, which addresses the CI-ICE task through two core aspects. Specifically, first, we propose a concept learning approach that leverages the inherent hierarchical modeling capability of hyperbolic space to achieve accurate concept disentanglement while preserving the hierarchical structure and relational dependencies among concepts; second, we introduce a concept-wise optimization method that maps the concept embedding space to maintain complex inter-concept relationships while ensuring concept composability. Our method demonstrates outstanding performance in extracting compositionally interpretable intrinsic concepts from a single image. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_11795 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Intrinsic Concept Extraction Based on Compositional Interpretability Shi, Hanyu Tao, Hong Huang, Guoheng Jiang, Jianbin Chen, Xuhang Pun, Chi-Man Wang, Shanhu Pan, Pan Computer Vision and Pattern Recognition Unsupervised Concept Extraction aims to extract concepts from a single image; however, existing methods suffer from the inability to extract composable intrinsic concepts. To address this, this paper introduces a new task called Compositional and Interpretable Intrinsic Concept Extraction (CI-ICE). The CI-ICE task aims to leverage diffusion-based text-to-image models to extract composable object-level and attribute-level concepts from a single image, such that the original concept can be reconstructed through the combination of these concepts. To achieve this goal, we propose a method called HyperExpress, which addresses the CI-ICE task through two core aspects. Specifically, first, we propose a concept learning approach that leverages the inherent hierarchical modeling capability of hyperbolic space to achieve accurate concept disentanglement while preserving the hierarchical structure and relational dependencies among concepts; second, we introduce a concept-wise optimization method that maps the concept embedding space to maintain complex inter-concept relationships while ensuring concept composability. Our method demonstrates outstanding performance in extracting compositionally interpretable intrinsic concepts from a single image. |
| title | Intrinsic Concept Extraction Based on Compositional Interpretability |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.11795 |