Concepts are complex, and often time contains abstract information to be distilled from real-world data. The problem of learning concepts in a complex and dynamic environment is essential to accomplish complex real-world missions. This topic has plausible real-world applications and is gathering much attention in the research community.
Here, we provide a non-exhaustive list of papers that study concept learning.
- Uncovering Unique Concept Vectors through Latent Space Decomposition [paper]
- Hierarchical Semantic Tree Concept Whitening for Interpretable Image Classification [paper]
- SHARCS: Shared Concept Space for Explainable Multimodal Learning [paper]
- A Holistic Approach to Unifying Automatic Concept Extraction and Concept Importance Estimation [paper]
- ConceptBed: Evaluating Concept Learning Abilities of Text-to-Image Diffusion Models [paper] [code]
- Mix-of-Show: Decentralized Low-Rank Adaptation for Multi-Concept Customization of Diffusion Models [paper]
- Concept Decomposition for Visual Exploration and Inspiration [paper]
- Explain Any Concept: Segment Anything Meets Concept-Based Explanation [paper]
- Causal Proxy Models for Concept-Based Model Explanations [paper] [code]
- ConceptLab: Creative Generation using Diffusion Prior Constraints [paper] [code]
- Unsupervised Compositional Concepts Discovery with Text-to-Image Generative Models (ICCV 2023) [paper] [code]
- Probabilistic Concept Bottleneck Models (ICML 2023) [paper] [code]
- Text-To-Concept (and Back) via Cross-Model Alignment (ICML 2023) [paper]
- Discover and Cure: Concept-aware Mitigation of Spurious Correlation (ICML 2023) [paper] [code]
- Interpretable Neural-Symbolic Concept Reasoning (ICML 2023) [paper] [code]
- Cones: Concept Neurons in Diffusion Models for Customized Generation (ICML 2023) [paper] [code]
- A Closer Look at the Intervention Procedure of Concept Bottleneck Models (ICML 2023) [paper] [code]
- Learning Bottleneck Concepts in Image Classification (CVPR 2023) [paper] [code]
- Dynamic Conceptional Contrastive Learning for Generalized Category Discovery (CVPR 2023) [paper] [code]
- Language in a Bottle: Language Model Guided Concept Bottlenecks for Interpretable Image Classification (CVPR 2023) [paper] [code]
- Label-Free Concept Bottleneck Models (ICLR 2023) [paper] [code]
- Actional Atomic-Concept Learning for Demystifying Vision-Language Navigation (AAAI 2023) [paper]
- Towards Robust Metrics for Concept Representation Evaluation (AAAI 2023) [paper]
- Multi-dimensional concept discovery (MCD): A unifying framework with completeness guarantees (TMLR 2023) [paper] [code]
- Concept Embedding Models: Beyond the Accuracy-Explainability Trade-Off (NeurIPS 2022) [paper] [code]
- ZeroC: A Neuro-Symbolic Model for Zero-shot Concept Recognition and Acquisition at Inference Time (NeurIPS 2022) [paper] [code]
- Interactive Disentanglement: Learning Concepts by Interacting with their Prototype Representations (CVPR 2022) [paper] [code]
- Concept Learning for Interpretable Multi-Agent Reinforcement Learning (CoRL 2022) [paper]
- beta-VAE: Learning Basic Visual Concepts with a Constrained Variational Framework (ICLR 2016) [paper] [code]
- Human-level concept learning through probabilistic program induction (Science 2015) [paper]
Please help us improve the above listing by submitting PRs of other papers in this space. Thank you!