English

Learning Concept Bottleneck Models from Mechanistic Explanations

Machine Learning 2026-03-10 v1 Artificial Intelligence

Abstract

Concept Bottleneck Models (CBMs) aim for ante-hoc interpretability by learning a bottleneck layer that predicts interpretable concepts before the decision. State-of-the-art approaches typically select which concepts to learn via human specification, open knowledge graphs, prompting an LLM, or using general CLIP concepts. However, concepts defined a-priori may not have sufficient predictive power for the task or even be learnable from the available data. As a result, these CBMs often significantly trail their black-box counterpart when controlling for information leakage. To address this, we introduce a novel CBM pipeline named Mechanistic CBM (M-CBM), which builds the bottleneck directly from a black-box model's own learned concepts. These concepts are extracted via Sparse Autoencoders (SAEs) and subsequently named and annotated on a selected subset of images using a Multimodal LLM. For fair comparison and leakage control, we also introduce the Number of Contributing Concepts (NCC), a decision-level sparsity metric that extends the recently proposed NEC metric. Across diverse datasets, we show that M-CBMs consistently surpass prior CBMs at matched sparsity, while improving concept predictions and providing concise explanations. Our code is available at https://github.com/Antonio-Dee/M-CBM.

Keywords

Cite

@article{arxiv.2603.07343,
  title  = {Learning Concept Bottleneck Models from Mechanistic Explanations},
  author = {Antonio De Santis and Schrasing Tong and Marco Brambilla and Lalana Kagal},
  journal= {arXiv preprint arXiv:2603.07343},
  year   = {2026}
}

Comments

ICLR 2026

R2 v1 2026-07-01T11:08:42.750Z