✨ Curated collection of papers and resources on latest advances of Concept based models. ✨
🗂️ Table of Contents
🚧🚧🚧 THIS IS UNDER CONSTRUCTION🚧🚧🚧
Do not hesitate to submit a pull request if you want to add a reference or any correction!
Standard CBM framework where the data is annotated by a human expert.
The framework combines a Visual-Language model to generate concept annotations.
Title | Focus | Venue | Date | Code |
---|---|---|---|---|
Measuring Leakage In Concept-Based Methods: An Information Theoretic Approach |
Leakage | ICLR | 2025 | - |
If Concept Bottlenecks are the Question, are Foundation Models the Answer? |
Concept Quality | 2025 | Github | |
Overlooked Factors in Concept-based Explanations: Dataset Choice, Concept Learnability, and Human Capability |
Interpretability | CVPR | 2023 | Github |
A closer look at the intervention procedure of concept bottleneck models |
Interventions | ICML | 2023 | Github |
IS DISENTANGLEMENT ALL YOU NEED? COMPARING CONCEPT-BASED & DISENTANGLEMENT APPROACHES |
Entanglement | ICLR | 2021 | Github |
Do Concept Bottleneck Models learn as intended |
Alignment | ICLR | 2021 | - |
Promises and Pitfalls of Black-Box Concept Learning Models |
Entanglement, Leakage | - | 2021 | - |
ON COMPLETENESS-AWARE CONCEPT-BASED EXPLANATIONS IN DEEP NEURAL NETWORKS |
Completeness | NeurIPS | 2020 | Github |
Github |