Concept-based models can use controlled 'benign' information leakage to remain accurate and intervenable under real-world concept incompleteness by reframing their training objective.
Is disentanglement all you need? comparing concept-based & disentanglement approaches
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
VH-CBM uses a Gaussian process in VLM embedding space to propagate sparse human annotations and improve concept accuracy and calibration over pure VLM-guided concept bottleneck models.
A VAE and slot-attention scheme learns interpretable object concepts from 1% labels, enabling symbolic reasoning that outperforms foundation models under domain shift.
Empirical tests of VLM-CBMs show VLM supervision differs from expert annotations depending on task and that concept accuracy correlates weakly with quality metrics.
citing papers explorer
-
In Defense of Information Leakage in Concept-based Models
Concept-based models can use controlled 'benign' information leakage to remain accurate and intervenable under real-world concept incompleteness by reframing their training objective.
-
Concepts Worth Having: Refining VLM-Guided Concept Bottleneck Models with Minimal Annotations
VH-CBM uses a Gaussian process in VLM embedding space to propagate sparse human annotations and improve concept accuracy and calibration over pure VLM-guided concept bottleneck models.
-
Weakly Supervised Concept Learning for Object-centric Visual Reasoning
A VAE and slot-attention scheme learns interpretable object concepts from 1% labels, enabling symbolic reasoning that outperforms foundation models under domain shift.
-
If Concept Bottlenecks are the Question, are Foundation Models the Answer?
Empirical tests of VLM-CBMs show VLM supervision differs from expert annotations depending on task and that concept accuracy correlates weakly with quality metrics.