# Concept Bottleneck Models Without Predefined Concepts

arXiv: https://arxiv.org/abs/2407.03921 · HTML: https://mlsanity.com/paper/2407.03921/

## Key facts

- Cited by 17 papers in the map (data through 14 Oct 2026).
- Builds on 34 of its 48 references that are in the map.
- First posted on arXiv 4 Jul 2024.

## Builds on (most cited first, 30 of 34)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Locating and Editing Factual Associations in GPT](https://mlsanity.com/paper/2202.05262/) (2022)
- [Visualizing and Understanding Convolutional Networks](https://mlsanity.com/paper/1311.2901/) (2013)
- [Sparse Autoencoders Find Highly Interpretable Features in Language Models](https://mlsanity.com/paper/2309.08600/) (2023)
- [Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)](https://mlsanity.com/paper/1711.11279/) (2017)
- [Concept Bottleneck Models](https://mlsanity.com/paper/2007.04612/) (2020)
- [Toy Models of Superposition](https://mlsanity.com/paper/2209.10652/) (2022)
- [Network Dissection: Quantifying Interpretability of Deep Visual Representations](https://mlsanity.com/paper/1704.05796/) (2017)
- [Erasing Concepts from Diffusion Models](https://mlsanity.com/paper/2303.07345/) (2023)
- [Visual Classification via Description from Large Language Models](https://mlsanity.com/paper/2210.07183/) (2022)
- [Language in a Bottle: Language Model Guided Concept Bottlenecks for Interpretable Image Classification](https://mlsanity.com/paper/2211.11158/) (2022)
- [Post-hoc Concept Bottleneck Models](https://mlsanity.com/paper/2205.15480/) (2022)
- [Label-Free Concept Bottleneck Models](https://mlsanity.com/paper/2304.06129/) (2023)
- [Modifying Memories in Transformer Models](https://mlsanity.com/paper/2012.00363/) (2020)
- [k-Sparse Autoencoders](https://mlsanity.com/paper/1312.5663/) (2013)
- [CRAFT: Concept Recursive Activation FacTorization for Explainability](https://mlsanity.com/paper/2211.10154/) (2022)
- [Promises and Pitfalls of Black-Box Concept Learning Models](https://mlsanity.com/paper/2106.13314/) (2021)
- [Invertible Concept-based Explanations for CNN Models with Non-negative Concept Activation Vectors](https://mlsanity.com/paper/2006.15417/) (2020)
- [Do Concept Bottleneck Models Learn as Intended?](https://mlsanity.com/paper/2105.04289/) (2021)
- [A Holistic Approach to Unifying Automatic Concept Extraction and Concept Importance Estimation](https://mlsanity.com/paper/2306.07304/) (2023)
- [Rewriting a Deep Generative Model](https://mlsanity.com/paper/2007.15646/) (2020)
- [Leveraging Sparse Linear Layers for Debuggable Deep Networks](https://mlsanity.com/paper/2105.04857/) (2021)
- [GlanceNets: Interpretabile, Leak-proof Concept-based Models](https://mlsanity.com/paper/2205.15612/) (2022)
- [Editing a classifier by rewriting its prediction rules](https://mlsanity.com/paper/2112.01008/) (2021)
- [Describing Differences between Text Distributions with Natural Language](https://mlsanity.com/paper/2201.12323/) (2022)
- [Describing Differences in Image Sets with Natural Language](https://mlsanity.com/paper/2312.02974/) (2023)
- [Multi-dimensional concept discovery (MCD): A unifying framework with completeness guarantees](https://mlsanity.com/paper/2301.11911/) (2023)
- [Overlooked factors in concept-based explanations: Dataset choice, concept learnability, and human capability](https://mlsanity.com/paper/2207.09615/) (2022)

## Built on by (most cited first, 17 of 17)

- [HypoBench: Towards Systematic and Principled Benchmarking for Hypothesis Generation](https://mlsanity.com/paper/2504.11524/) (2025)
- [Interpretable and Testable Vision Features via Sparse Autoencoders](https://mlsanity.com/paper/2502.06755/) (2025)
- [Explaining Datasets in Words: Statistical Models with Natural Language Parameters](https://mlsanity.com/paper/2409.08466/) (2024)
- [DCBM: Data-Efficient Visual Concept Bottleneck Models](https://mlsanity.com/paper/2412.11576/) (2024)
- [If Concept Bottlenecks are the Question, are Foundation Models the Answer?](https://mlsanity.com/paper/2504.19774/) (2025)
- [Shortcuts and Identifiability in Concept-based Models from a Neuro-Symbolic Lens](https://mlsanity.com/paper/2502.11245/) (2025)
- [Predictive Concept Decoders: Training Scalable End-to-End Interpretability Assistants](https://mlsanity.com/paper/2512.15712/) (2025)
- [Object Centric Concept Bottlenecks](https://mlsanity.com/paper/2505.24492/) (2025)
- [Dataset Featurization: Uncovering Natural Language Features through Unsupervised Data Reconstruction](https://mlsanity.com/paper/2502.17541/) (2025)
- [Adaptive Test-Time Intervention for Concept Bottleneck Models](https://mlsanity.com/paper/2503.06730/) (2025)
- [Explainable deep learning improves human mental models of self-driving cars](https://mlsanity.com/paper/2411.18714/) (2024)
- [COMIX: Compositional Explanations using Prototypes](https://mlsanity.com/paper/2501.06059/) (2025)
- [ImageSet2Text: Describing Sets of Images through Text](https://mlsanity.com/paper/2503.19361/) (2025)
- [Learning reusable concepts across different egocentric video understanding tasks](https://mlsanity.com/paper/2505.24690/) (2025)
- [Concept Inconsistency in Dermoscopic Concept Bottleneck Models: A Rough-Set Analysis of the Derm7pt Dataset](https://mlsanity.com/paper/2604.19323/) (2026)
- [Measuring What Matters: Synthetic Benchmarks for Concept Bottleneck Models](https://mlsanity.com/paper/2606.04326/) (2026)
- [Concept Flow Models: Anchoring Concept-Based Reasoning with Hierarchical Bottlenecks](https://mlsanity.com/paper/2606.19489/) (2026)

## Abstract (arXiv)

There has been considerable recent interest in interpretable concept-based models such as Concept Bottleneck Models (CBMs), which first predict human-interpretable concepts and then map them to output classes. To reduce reliance on human-annotated concepts, recent works have converted pretrained black-box models into interpretable CBMs post-hoc. However, these approaches predefine a set of concepts, assuming which concepts a black-box model encodes in its representations. In this work, we eliminate this assumption by leveraging unsupervised concept discovery to automatically extract concepts without human annotations or a predefined set of concepts. We further introduce an input-dependent concept selection mechanism that ensures only a small subset of concepts is used across all classes. We show that our approach improves downstream performance and narrows the performance gap to black-box models, while using significantly fewer concepts in the classification. Finally, we demonstrate how large vision-language models can intervene on the final model weights to correct model errors.