mlsanityData to 13 Oct 2026

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

arXiv 2502.01191 ↗First posted 3 Feb 2025cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. No citing papers in the map yet (data through 13 Oct 2026). Builds on 18 of its 26 references that are in the map. First posted on arXiv 3 Feb 2025.

Builds on

18 of 26 references.

References · 18

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,950
  2. 2014Deep Learning Face Attributes in the Wildcited by 3,686
  3. 2020Concept Bottleneck Modelscited by 1,027
  4. 2016InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Netscited by 860
  5. 2017Zero-Shot Learning -- A Comprehensive Evaluation of the Good, the Bad and the Uglycited by 538
  6. 2018Towards a Definition of Disentangled Representationscited by 295
  7. 2022Post-hoc Concept Bottleneck Modelscited by 292
  8. 2021Promises and Pitfalls of Black-Box Concept Learning Modelscited by 131
  9. 2021Do Concept Bottleneck Models Learn as Intended?cited by 116
  10. 2018Robustly Disentangled Causal Mechanisms: Validating Deep Representations for Interventional Robustnesscited by 108
  11. 2023Probabilistic Concept Bottleneck Modelscited by 101
  12. 2022Concept Embedding Models: Beyond the Accuracy-Explainability Trade-Offcited by 61
6 more
  1. 2021Multi-VAE: Learning Disentangled View-common and View-peculiar Visual Representations for Multi-view Clusteringcited by 51
  2. 2023Auxiliary Losses for Learning Generalizable Concept-based Modelscited by 35
  3. 2019OOGAN: Disentangling GAN with One-Hot Sampling and Orthogonal Regularizationcited by 17
  4. 2020Improving Robustness and Generality of NLP Models Using Disentangled Representationscited by 8
  5. 2023Towards a Deeper Understanding of Concept Bottleneck Models Through End-to-End Explanationcited by 7
  6. 2024Disentangling Instructive Information from Ranked Multiple Candidates for Multi-Document Scientific Summarizationcited by 2

Built on by

Not cited yet.