Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2507.08334 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908860809216000 |
|---|---|
| author | Kim, Sangwon Lee, Kyoungoh Dong, Jeyoun Kim, Kwang-Ju |
| author_facet | Kim, Sangwon Lee, Kyoungoh Dong, Jeyoun Kim, Kwang-Ju |
| contents | Generative concept bottleneck models aim to enable interpretable generation by routing synthesis through explicit, user-facing concepts. In practice, prior approaches often rely on non-explicit bottleneck representations (e.g., vision cues or opaque concept embeddings) or black-box decoders to preserve image quality, which weakens the transparency. We propose CoBELa (Concept Bottlenecks on Energy Landscapes), a decoder-free, energy-based framework that eliminates non-explicit bottleneck representations by conditioning generation entirely through per-concept energy functions over the latent space of a frozen pretrained generator-requiring no generator retraining and enabling post-hoc interpretation. Because these concept energies compose additively, CoBELa naturally supports compositional concept interventions: concept conjunction and negation are realized by summing or subtracting per-concept energy terms without additional training. A diffusion-scheduled energy guidance scheme further replaces expensive MCMC chains with more stable, scheduled denoising for efficient concept-steered sampling. Experiments on CelebA-HQ and CUB-200-2011 demonstrate improvements over prior concept bottleneck generative models, achieving 75.70%/82.42% concept accuracy and 6.47/5.37 FID, respectively, while enabling reliable multi-concept interventions. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_08334 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | CoBELa: Steering Transparent Generation via Concept Bottlenecks on Energy Landscapes Kim, Sangwon Lee, Kyoungoh Dong, Jeyoun Kim, Kwang-Ju Computer Vision and Pattern Recognition Artificial Intelligence Generative concept bottleneck models aim to enable interpretable generation by routing synthesis through explicit, user-facing concepts. In practice, prior approaches often rely on non-explicit bottleneck representations (e.g., vision cues or opaque concept embeddings) or black-box decoders to preserve image quality, which weakens the transparency. We propose CoBELa (Concept Bottlenecks on Energy Landscapes), a decoder-free, energy-based framework that eliminates non-explicit bottleneck representations by conditioning generation entirely through per-concept energy functions over the latent space of a frozen pretrained generator-requiring no generator retraining and enabling post-hoc interpretation. Because these concept energies compose additively, CoBELa naturally supports compositional concept interventions: concept conjunction and negation are realized by summing or subtracting per-concept energy terms without additional training. A diffusion-scheduled energy guidance scheme further replaces expensive MCMC chains with more stable, scheduled denoising for efficient concept-steered sampling. Experiments on CelebA-HQ and CUB-200-2011 demonstrate improvements over prior concept bottleneck generative models, achieving 75.70%/82.42% concept accuracy and 6.47/5.37 FID, respectively, while enabling reliable multi-concept interventions. |
| title | CoBELa: Steering Transparent Generation via Concept Bottlenecks on Energy Landscapes |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2507.08334 |