Papers
arxiv:2507.23784

SUB: Benchmarking CBM Generalization via Synthetic Attribute Substitutions

Published on Jul 31
Authors:
,
,
,

Abstract

A novel benchmark and Tied Diffusion Guidance method assess the robustness of Concept Bottleneck Models to concept variations under distribution shifts.

AI-generated summary

Concept Bottleneck Models (CBMs) and other concept-based interpretable models show great promise for making AI applications more transparent, which is essential in fields like medicine. Despite their success, we demonstrate that CBMs struggle to reliably identify the correct concepts under distribution shifts. To assess the robustness of CBMs to concept variations, we introduce SUB: a fine-grained image and concept benchmark containing 38,400 synthetic images based on the CUB dataset. To create SUB, we select a CUB subset of 33 bird classes and 45 concepts to generate images which substitute a specific concept, such as wing color or belly pattern. We introduce a novel Tied Diffusion Guidance (TDG) method to precisely control generated images, where noise sharing for two parallel denoising processes ensures that both the correct bird class and the correct attribute are generated. This novel benchmark enables rigorous evaluation of CBMs and similar interpretable models, contributing to the development of more robust methods. Our code is available at https://github.com/ExplainableML/sub and the dataset at http://huggingface.co/datasets/Jessica-bader/SUB.

Community

Sign up or log in to comment

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2507.23784 in a model README.md to link it from this page.

Datasets citing this paper 1

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2507.23784 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.