Towards Robust and Reliable Concept Representations: Reliability-Enhanced Concept Embedding Model

TL;DR

This paper introduces RECEM, a novel model that improves the reliability and robustness of concept representations in concept bottleneck models by disentangling irrelevant features and aligning semantics across samples.

Contribution

RECEM is the first to combine concept-level disentanglement with a concept mixup mechanism to enhance concept reliability and robustness in CBMs.

Findings

RECEM outperforms baselines on multiple datasets.

It maintains performance under background and domain shifts.

Demonstrates improved interpretability and stability.

Abstract

Concept Bottleneck Models (CBMs) aim to enhance interpretability by predicting human-understandable concepts as intermediates for decision-making. However, these models often face challenges in ensuring reliable concept representations, which can propagate to downstream tasks and undermine robustness, especially under distribution shifts. Two inherent issues contribute to concept unreliability: sensitivity to concept-irrelevant features (e.g., background variations) and lack of semantic consistency for the same concept across different samples. To address these limitations, we propose the Reliability-Enhanced Concept Embedding Model (RECEM), which introduces a two-fold strategy: Concept-Level Disentanglement to separate irrelevant features from concept-relevant information and a Concept Mixup mechanism to ensure semantic alignment across samples. These mechanisms work together to improve concept reliability, enabling the model to focus on meaningful object attributes and generate faithful concept representations. Experimental results demonstrate that RECEM consistently outperforms existing baselines across multiple datasets, showing superior performance under background and domain shifts. These findings highlight the effectiveness of disentanglement and alignment strategies in enhancing both reliability and robustness in CBMs.