# Teaching Compositionality to CNNs

arXiv: https://arxiv.org/abs/1706.04313 · HTML: https://mlsanity.com/paper/1706.04313/

## Key facts

- Cited by 18 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 Jun 2017.

## Built on by (most cited first, 18 of 18)

- [AGQA: A Benchmark for Compositional Spatio-Temporal Reasoning](https://mlsanity.com/paper/2103.16002/) (2021)
- [Learning Disentangled Representations in the Imaging Domain](https://mlsanity.com/paper/2108.12043/) (2021)
- [Compositional Generalization in Unsupervised Compositional Representation Learning: A Study on Disentanglement and Emergent Language](https://mlsanity.com/paper/2210.00482/) (2022)
- [EXplainable Neural-Symbolic Learning (X-NeSyL) methodology to fuse deep learning representations with expert knowledge graphs: the MonuMAI cultural heritage use case](https://mlsanity.com/paper/2104.11914/) (2021)
- [Learning Compositional Representation for 4D Captures with Neural ODE](https://mlsanity.com/paper/2103.08271/) (2021)
- [CX-ToM: Counterfactual Explanations with Theory-of-Mind for Enhancing Human Trust in Image Recognition Models](https://mlsanity.com/paper/2109.01401/) (2021)
- [Interpretable Compositional Convolutional Neural Networks](https://mlsanity.com/paper/2107.04474/) (2021)
- [Deep Graph Convolutional Reinforcement Learning for Financial Portfolio Management -- DeepPocket](https://mlsanity.com/paper/2105.08664/) (2021)
- [Does Data Scaling Lead to Visual Compositional Generalization?](https://mlsanity.com/paper/2507.07102/) (2025)
- [Greybox XAI: a Neural-Symbolic learning framework to produce interpretable predictions for image classification](https://mlsanity.com/paper/2209.14974/) (2022)
- [Compositionally Equivariant Representation Learning](https://mlsanity.com/paper/2306.07783/) (2023)
- [Compositional Learning of Visually-Grounded Concepts Using Reinforcement](https://mlsanity.com/paper/2309.04504/) (2023)
- [The Construction of Reality in an AI: A Review](https://mlsanity.com/paper/2302.05448/) (2023)
- [AI-based association analysis for medical imaging using latent-space geometric confounder correction](https://mlsanity.com/paper/2311.12836/) (2023)
- [Two-stage Vision Transformers and Hard Masking offer Robust Object Representations](https://mlsanity.com/paper/2506.08915/) (2025)
- [Enhancing Cross-Modal Medical Image Segmentation through Compositionality](https://mlsanity.com/paper/2408.11733/) (2024)
- [Human-like compositional learning of visually-grounded concepts using synthetic environments](https://mlsanity.com/paper/2504.06618/) (2025)
- [Let's Split Up: Zero-Shot Classifier Edits for Fine-Grained Video Understanding](https://mlsanity.com/paper/2602.16545/) (2026)

## Abstract (arXiv)

Convolutional neural networks (CNNs) have shown great success in computer vision, approaching human-level performance when trained for specific tasks via application-specific loss functions. In this paper, we propose a method for augmenting and training CNNs so that their learned features are compositional. It encourages networks to form representations that disentangle objects from their surroundings and from each other, thereby promoting better generalization. Our method is agnostic to the specific details of the underlying CNN to which it is applied and can in principle be used with any CNN. As we show in our experiments, the learned representations lead to feature activations that are more localized and improve performance over non-compositional baselines in object recognition tasks.