Abstract
Caption Bottleneck Models use free-form captions generated by large multimodal models and a text classifier to produce interpretable, leakage-free predictions while automatically discovering dataset-specific concepts.
Concept Bottleneck Models (CBMs) provide interpretability by routing predictions through a layer of human-understandable concepts. However, defining an optimal concept set for a specific dataset remains an open challenge. Existing approaches rely on expensive expert annotations or LLM-generated lists based solely on class names. Even "open-vocabulary" variants typically depend on static concept sets, which restrict discovery and introduce label bias. Furthermore, traditional CBMs often suffer from information leakage, where unmodeled visual features bypass the bottleneck and compromise the integrity of the explanations. To overcome these limitations, we propose Caption Bottleneck Models (CaBM), a framework that circumvents the need for predefined concept sets by replacing rigid concept layers with free-form natural language. By representing images via LMM-generated captions and training a classifier strictly on this text, CaBM ensures a leakage-free architecture by construction. Additionally, by analyzing the text classifier post-training, CaBM autonomously discovers high-quality, dataset-specific concepts. Our results across fine- and coarse-grained benchmarks demonstrate that CaBM achieves competitive accuracy while preserving interpretability without the constraints of external dictionaries or manual labeling.
Get this paper in your agent:
hf papers read 2607.00578 Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash Models citing this paper 6
bariscagliyan/cabm-cifar100
Datasets citing this paper 1
bariscagliyan/cabm-captions
Spaces citing this paper 0
No Space linking this paper
Collections including this paper 0
No Collection including this paper