aiqbench analysis
This study empirically validates that Concept Bottleneck Models (CBMs) enhance human-AI team accuracy specifically when tasks are difficult, concepts are clear, and users actively intervene with the model's logic. The findings highlight a critical dependency on accurate concept detection, as errors in this area can erode user trust rather than aid decision-making. These results offer essential practical guidelines for deploying interpretable AI architectures in real-world collaborative workflows where human oversight is required.
Are Concept Bottleneck Models Effective as Decision-Support Systems?
arXiv:2608.25581v1 Announce Type: cross Abstract: Concept Bottleneck Models (CBMs) are interpretable-by-design neural networks that detect human-understandable concepts from the input and use them to generate predictions. By allowing users to inspect the concepts underlying a prediction and explore how predictions change under alternative concept configurations, CBMs have emerged as one of the most prominent approaches to supporting human-AI collaboration. However, user studies investigating their actual effectiveness as decision-support systems remain limited. We present two large-scale user studies (N participants = 705, N observations = 6,959) evaluating how concept-based explanations and user interventions on the model's concepts affect the performance of the human-AI team in two distinct binary classification tasks. Our results show that CBMs, and particularly their interactive component, can improve human-AI team accuracy relative to both unaided human performance and performance with non-interpretable AI support. However, these benefits emerge only under certain conditions: classification tasks perceived as difficult, easily identifiable concepts, and active interaction with the model. We also discuss how inaccurate concept detection may undermine users' trust in the model. Overall, this work provides practical guidance for the deployment of CBMs as effective decision-support tools.
Title:Are Concept Bottleneck Models Effective as Decision-Support Systems?
Abstract:Concept Bottleneck Models (CBMs) are interpretable-by-design neural networks that detect human-understandable concepts from the input and use them to generate predictions. By allowing users to inspect the concepts underlying a prediction and explore how predictions change under alternative concept configurations, CBMs have emerged as one of the most prominent approaches to supporting human-AI collaboration. However, user studies investigating their actual effectiveness as decision-support systems remain limited. We present two large-scale user studies (N participants = 705, N observations = 6,959) evaluating how concept-based explanations and user interventions on the model's concepts affect the performance of the human-AI team in two distinct binary classification tasks. Our results show that CBMs, and particularly their interactive component, can improve human-AI team accuracy relative to both unaided human performance and performance with non-interpretable AI support. However, these benefits emerge only under certain conditions: classification tasks perceived as difficult, easily identifiable concepts, and active interaction with the model. We also discuss how inaccurate concept detection may undermine users' trust in the model. Overall, this work provides practical guidance for the deployment of CBMs as effective decision-support tools.
Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
arXiv:2608.25581 [cs.HC]
(or arXiv:2608.25581v1 [cs.HC] for this version)
https://doi.org/10.48550/arXiv.2608.25581 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 285 words
Related coverage
[AINews] High Taste Testers Ep. 1
Latent Space · Oct 17, 2026
[AINews] Opus 5.5 is good at explainer videos
Latent Space · Sep 29, 2026
State of the Field: AI for Epistemics and Coordination
LessWrong · Oct 2, 2026
[AINews] The Future of Latent Space
Latent Space · Sep 25, 2026